Noun Class Extraction from a Corpus-based Collocation Dictionary-Chu-Ren Huang
a_corpus-based_study_of_the_use_of_MAKE_in_Chinese_student_writing(2)
II. Literature Review
What is Corpus?
A collection of naturally occurring language text chosen to characterize a state or variety of a language. John Sinclair (1991) A collection of naturally occurring examples of language, consisting of anything from a few sentences to a set of written texts or tape recordings, which have been collected for linguistic study. Susan Hunston (2002) Learner Corpus: systematic computerized collections of texts produced by learners. Nadja Nesselhauf (2004)
Teaching data-driven learning
Pedagogic material Nadja Nesselhauf (2004)
7
Studies on MAKE
• Altenberg & Granger (2001):
– Corpus:
• Learner corpus: International Corpus of Learner English (ICLE) (advance learners)
2
I. Introduction
• Chinese learners’ use of MAKE: This paper focus on the use of MAKE in the writing of university students in China and in Taiwan by the Chinese Learner English Corpus (CLEC) and the Soochow Colber Student Corpus (SCSC). • Benchmark: BNC vs. LOCNESS
名词解释
According to "The Canadian Modern Language Review", formulaic sequences are "fixed combinations of words that...can facilitate fluency in speech by making pauses shorter and less frequent, and allowing longer runs of speech between pauses".[1]
the comparative of many disyllabic adjectives can be formed either with the word more or with the suffix -er, for example more stupid or stupider
目前DJ音标符号共计有48个，其中元音有20个﹔辅音有28个。
DJ音标初期可能是为了节省印刷模具数量，只有11个符号不同于英文字母；后来发明者的学生为了一符一音，变为有15个符号不同于英文字母。
The Prague school, or Prague linguistic circle,[1] was an influential[2] group of literary critics and linguists in Prague. Its proponents developed methods of structuralist literary analysis[3] and a theory of the standard language and of language cultivation during the years 1928–1939. The linguistic circle was founded in the Café Derby in Prague, which is also where meetings took place during its first years.[4]
关系抽取研究综述
关系抽取研究综述母克东;万琪【摘要】信息抽取、自然语言理解、信息检索等应用需要更好地理解两个实体之间的语义关系，对关系抽取进行概况总结。
将关系抽取划分为两个阶段研究：特定领域的传统关系抽取和开放领域的关系抽取。
并对关系抽取从抽取算法、评估指标和未来发展趋势三个部分对关系抽取系统进行系统的分析总结。
%Many applications in natural language understanding, information extraction, information retrieval require an understanding of the seman-tic relations between entities. Carries on the summary to the relation extraction. There are two paradigms extracting the relation-ship be-tween two entities: the Traditional Relation Extraction and the Open Relation Extraction. Makes detailed introduction and analysis of the algorithm of relation extraction, evaluation indicators and the future of the relation extraction system.【期刊名称】《现代计算机（专业版）》【年(卷),期】2015(000)002【总页数】4页(P18-21)【关键词】关系抽取;机器学习;信息抽取;开放关系抽取【作者】母克东;万琪【作者单位】四川大学计算机学院，成都 610065;四川大学计算机学院，成都610065【正文语种】中文随着大数据的不断发展，海量信息以半结构或者纯原始文本的形式展现给信息使用者，如何采用自然语言处理和数据挖掘相关技术从中帮助用户获取有价值的信息，是当代计算机研究技术迫切的需求。
corpus-based vocabulary teaching
English Lexical Studies
• John Sinclair • established in Edinburgh in Birmingham in 1963 • contained a small electronic sample of spoken and written language with a focus of study on he meaning of lexical items, including collocations.
Corpus-based vocabulary teaching
corpus
vocabulary teaching
corpus
Corpus (singular), corpora (plural) in Latin means “body” In modern linguistics, a corpus can be defined as a body of naturally occurring language. “A corpus is nothing more nor less than a collection of texts input into a computer” (Fox, 1998, p. 25)
corpus
„A corpus is a collection of pieces of language that are selected and ordered according to explicit linguistic criteria in order to be used as a sample of the language.‟ (Sinclair, 1996) It should be added that computer corpora are rarely haphazard collections of textual material: They are generally assembled with particular purposes in mind, and are often assembled to be (informally speaking) representative of some language or text type. (Leech 1992: 116)
语料库基本知识
.
6
计算语言学
◦ “计算语言学是研究用机器来处理自然语言的学科。它是由信息技
术和语言学交叉而成的”（CuS：1）。SLP没有直接提出计算语言 学的确切定义。SLP的作者在开篇借用了Stanley Kubrick科幻片中 的人物HAL，HAL是一个通晓英语的机器人。作者引入HAL的目 的在于说明，为了构建这样一个可与人通过自然语言进行交流的机 器人，需要哪些知识和技术：语言理解方面有语音识别和自然语言 理解（包括唇读技术），表达方面需要自然语言生成和语音合成， 另外HAL也需要信息检索、信息提取和推理方面的技能。而解决这 些问题一般涉及以下学科：自然语言处理，计算语言学，语音识别 和合成。SLP的作者将这三者合起来称为语音及语言处理，除了以 上HAL所用的这些技能外，SLP也囊括了其他重要的语言处理领域， 如：拼写校正、语法检查和机器翻译。
.
42
语言设定
.
43
.
44
(如逗号、句号等) 包括在内，但这一点有例外，如数字3.1415925 和整数的千分位分隔符(如100，000) 中的逗号等。
为了便于统计，对英语进行分词时通常在以上我们所说的“ 形符” 后加空格，使得他们与文本中的其他形符或符号分离开来。
.
11
类符(type)作为一个统计量，指语料库文本中任何一个独特的词形(word form)。换言之，在一个文本中，重复出现的形符只能记作一个类符。
◦ 都可以对语言学的语音、词汇、句法和语义等层面进行
统计和研究。
.
8
联系： ◦ 统计语言学和计量语言学都是利用统计方法来实现对语言成
分的统计，计量语言学以发现语言成分或语言成分间的数学 规律为目标。而统计语言学以所统计的语言特征在统计学上 显著和不显著为目标。
nat英文用法
nat英文用法Natural language processing (NLP) is a subfield of artificial intelligence that focuses on the interaction between humans and computers using natural language. It involves the development of algorithms and models that enable computers to understand and comprehend human language in a way that is similar to how humans do.NLP has a wide range of applications and is used in various domains such as automated customer support, chatbots, machine translation, sentiment analysis, text summarization, and information extraction. It has become an essential component of many modern technologies and has revolutionized the way humans interact with computers.One of the main challenges in NLP is understanding the ambiguity and complexity of human language. Natural language is inherently ambiguous, and words can have multiple meanings depending on the context in which they are used. For example, the word "bank" can refer to a financial institution or the edge of a river. NLP algorithms aim to disambiguate these words by analyzing the surrounding context and using statistical models to determine the most likely meaning.One common technique used in NLP is called part-of-speech tagging. It involves labeling each word in a sentence with its corresponding part of speech, such as noun, verb, adjective, or adverb. This information is crucial for understanding the grammatical structure of a sentence and for performing more advanced tasks such as syntactic parsing and semantic role labeling.Another important aspect of NLP is named entity recognition (NER). NER aims to identify and classify named entities in a text, such as person names, organizations, locations, and dates. This information is useful in applications like information extraction, where specific entities need to be identified and extracted from a larger text corpus.Sentiment analysis is another popular application of NLP. It involves the use of machine learning models to classify the sentiment or emotional tone of a given text. Sentiment analysis can be used to analyze customer reviews, social media posts, or political speeches to gain insights into public opinion or to identify trends and patterns.Machine translation is perhaps one of the most well-known applications of NLP. It involves automatically translating text from one language to another. Machine translation systems use various techniques, such as statistical models or neural networks, to learn the mapping between different languages and to generate accurate and fluent translations.Text summarization is another NLP task that involves automatically generating summaries of larger texts. This can be useful in situations where there is a large volume of information and a need to extract only the most important and relevant points. Text summarization algorithms can use techniques like extractive summarization, where important sentences or phrases are selected from the original text, or abstractive summarization, where a summary is generated by generating new sentences that capture theessence of the original text.Overall, NLP has made significant progress in recent years and has become an invaluable tool in many industries. With the advancements in machine learning and artificial intelligence, NLP algorithms will continue to improve, allowing computers to understand and generate human language more accurately and effectively. It is an exciting field with endless possibilities and is set to shape the future of human-computer interaction.。
Ethanol extracts frommerocallis citrina attenuate the decreases of brain-derived neurotrophic factor
Ethanol extracts from Hemerocallis citrina attenuate the decreasesof brain-derived neurotrophic factor,TrkB levels in rat induced by corticosterone administrationLi-Tao Yi a,n,Jing Li a,Huo-Chen Li b,Ying Zhou b,Bi-Fang Su a,Ke-Feng Yang b,Meng Jiang b,Yan-Ting Zhang aa Department of Chemical and Pharmaceutical Engineering,College of Chemical Engineering,Huaqiao University,Xiamen361021,Fujian province,PR Chinab Department of Biotechnology and Bioengineering,College of Chemical Engineering,Huaqiao University,Xiamen361021,Fujian province,PR Chinaa r t i c l e i n f oArticle history:Received5April2012Received in revised form27August2012Accepted6September2012Available online18September2012Keywords:Hemerocallis citrinaAntidepressant-like effectCorticosteroneBrain-derived neurotrophic factorTrkBa b s t r a c tEthnopharmacological relevance:Hemerocallis citrina,a traditional herbal medicine,has been used forthe improvement of behavioral and emotional status in Eastern-Asia countries.Aim of the study:Our previous studies have demonstrated that the ethanol extracts of H.citrinaflowers(HCE)reversed the behavioral alterations and monoamine neurotransmitter dysfunctions in stressedmice.However,the relation of its antidepressant-like action with neurotrophic molecular expressionsremains unknown.Materials and methods:To clarify this,we explored the effect of HCE(32.5,65,130mg/kg,p.o.)on thebehavior,brain-derived neurotrophic factor(BDNF)and its receptor(TrkB)in depression-like ratsinduced by exogenous administration of the stress hormone corticosterone(40mg/kg,s.c.).Results:It was observed that repeated administration of corticosterone induced an elevation on theserum corticosterone levels,which caused the abnormalities observed in the sucrose preference testand forced swimming test(FST).Administration of HCE(65and130mg/kg)reversed the changes aboveand up-regulated the BDNF and TrkB receptor protein expressions in the brain region of frontal cortexand hippocampus.Conclusion:Thesefindings confirm that HCE produce an antidepressant-like effect in corticosterone-induced depression-like model of rats and this effect is at least partly mediated by BDNF-TrkB signalingin the frontal cortex and hippocampus.&2012Elsevier Ireland Ltd.All rights reserved.1.IntroductionHemerocallis citrina Baroni(Liliaceae)which is widely grown inChina,Japan and Korea has been used for food and medicinalpurposes for thousands of years.According to the phytochemicaland pharmacological studies,H.citrina mainly containsflavo-noids,polyphenols and essential oils that contribute to the herb’sbiological effects,such as antibacterial(Zhan et al.,2005),anti-oxidant(Lang and Luo,2007)and nitrite-eliminating(Fu et al.,2009)activity.In addition,the effects of H.citrina such asrelieving gloom and improving sleeping were recorded in thefamous pharmaceutical text of‘‘Compendium of Materia Medica’’,the most complete and comprehensive medical book ever writtenin the history of traditional Chinese medicine.Clinically,H.citrinahas been widely used for the treatment of depressive disorders inChina(Chen et al.,2008).In previous studies,we found that the ethanol extracts ofH.citrina(HCE)administration significantly reduced the immo-bility time in both the forced swimming test(FST)and tailsuspension test(TST)without accompanying changes in locomo-tor activity in the open-field test(OFT)in mice(Gu et al.,2012).Furthermore,we also demonstrated that HCE enhanced serotonin(5-HT)and noradrenaline(NA)levels in the frontal cortex andhippocampus as well as elevated dopamine(DA)levels in thefrontal cortex(Gu et al.,2012).However,until now,the neuro-trophic molecular expression underlying the antidepressant-likeeffect of HCE remains unclear.Neurotrophic factors are critical regulators of the formationand plasticity of neuronal networks(Lee and Kim,2010).Brain-derived neurotrophic factor(BDNF),one of important neuro-trophic factors,has also been implicated in the etiology of majorContents lists available at SciVerse ScienceDirectjournal homepage:/locate/jepJournal of Ethnopharmacology0378-8741/$-see front matter&2012Elsevier Ireland Ltd.All rights reserved./10.1016/j.jep.2012.09.016Abbreviations:ANOVA,analysis of variance;BDNF,brain-derived neurotrophicfactor;DA,dopamine;FST,forced swimming test;GR,glucocorticoid receptor;HCE,ethanol extracts of Hemerocallis citrina;NA,noradrenaline;OFT,open-fieldtest;5-HT,serotonin;TST,tail suspension testn Corresponding author.Tel./fax:þ865926162302.E-mail addresses:litao.yi@,litaoyi@(L.-T.Yi).Journal of Ethnopharmacology144(2012)328–334depression and the mechanism of antidepressant treatment (Numakawa et al.,2010).Clinical study demonstrates that depres-sion is associated with reduced brain BDNF and its receptor (TrkB)levels and that the reductions can be restored up to the normal value by antidepressant treatment (Thompson et al.,2011).Animal studies have also demonstrated clear evidence that BDNF signaling through TrkB is involved in the mechanisms of action of antidepressant agents (Wang et al.,2010;Kutiyanawalla et al.,2011).BDNF and TrkB levels can therefore be useful markers for antidepressant-like response.In the present study,we used a rat depression model by repeated corticosterone injections to investigate the antidepressant-like effect of HCE.Furthermore,the study explored whether the HCE could reverse the loss of BDNF and its receptor TrkB mRNA and protein levels in the rat brain regions of frontal cortex and hippocampus.2.Materials and methods 2.1.AnimalsMale Sprague-Dawley rats (220–250g)were purchased from Laboratory Animal Center,Fujian Medical University,Fujian Province,PR China.Animals were singly housed under a normal 12-h/12-h light/dark schedule with the lights on at 07:00a.m.and had free access to tap water and food pellets.Ambient tempera-ture and relative humidity were maintained at 22721C and at 5575%,and given a standard chow and water ad libitum for the duration of the study.The animals were allowed 1week to acclimatize themselves to the housing conditions before the beginning of the experiments.All procedures were performed in accordance with the published guidelines of the China Council on Animal Care (Regulations for the Administration of Affairs Con-cerning Experimental Animals,approved by the State Council on October 31,1988and promulgated by Decree No.2of the State Science and Technology Commission on November 14,1988).2.2.Chemicals and reagentsCorticosterone was purchased from TCI Development Co.,Ltd.(Tokyo,Japan).Fluoxetine hydrochloride was purchased from Changzhou Siyao Pharmaceuticals Co.,Ltd.(Changzhou,PR China).All primers used in this study were designed and synthe-sized by Sangon Biotech Co.Ltd.(Shanghai,PR China).The anti-BDNF and anti-TrkB antibody and the respective secondary antibody were purchased from Santa Cruz Biotechnology Inc.(Santa Cruz,USA).Trizol reagent was purchased from Invitrogen (Carlsbad,USA).Reverse transcriptase Moloney Murine Leukemia Virus (M-MLV)used for cDNA synthesis was from Promega Corporation (Madison,USA).All other reagents used in RT-PCR and western bolt were purchased from Sangon Biotech Co.Ltd.(Shanghai,PR China).2.3.Plant material and HPLC analysisFlowers of H.citrina Baroni (Liliaceae)was purchased from Xiamen Wal-Mart and authenticated by Cheng-Fu Li,Department of Pharmacy,Xiamen Hospital of Traditional Chinese medicine (Voucher specimen number HU/CE-04281).Since rutin is the main component of H.citrina (Yang et al.,2006),a fingerprint was analyzed in HPLC using rutin as an index.A Shimadzu Prominence LC-20A HPLC system equipped with an UV detector was used.Qualitative and quantitative analyses of rutin were performed using an Agilent C 18column (4.6mM Â250mM,5m m),and the column temperatures were kept at 251C.A linear gradient elution profile was used in our study [0–20min:methanol to 0.5%acetic acid water solution (v/v)ratio of 20/80to 30/70;20–30min:30/70;30–70min:30/70to 80/20].Flow rates of elution in both cases were 1ml/min.The injection volume for the analyte was 10m l.The detection wavelength was set at 254nm for rutin.The raw material H.citrina was extracted with methanol at 251C for 30min.The content of the marker sub-stance rutin in H.citrina was calculated as 0.39%(Fig.1A and B).2.4.Preparation of the ethanol extractsDried sample (500g)was cut into small pieces and extracted three times in a reflux condenser for 4h with 3L of 75%ethanol by sonication at room temperature (25721C).The solutions were combined,filtered,concentrated under reduced pressure and lyophilized into powders.The final yield was 7.65%(w/w).2.5.Drug treatmentsDifferent groups of rats,7animals per group,were used for drug treatment and tests.Doses were calculated as mg/kg body weight.All the experimental animals including control and drug-treatment groups were simultaneously deprived food but not water 1h prior to drug administration.The aim of food with-drawn prior to drug administration was just to ensure the bioavailability of drug.In other time periods,all animals had free access to food.The standard dose of HCE in rat (130mg/kg)was based on our previous mouse study (Gu et al.,2012)and calculated on the basis of body surface area ratio between rat and mouse.In addition,considering the long-term treatment used in the present study,the doses of 32.5,65,130mg/kg every 24h were selected.The procedure and dose of corticosterone administration was performed as described in Mao et al.(2012).Corticosterone (40mg/kg),which was dissolved in a saline solution containing 0.1%dimethyl sulfoxide and 0.1%Tween-80,was administrated by subcutaneously (s.c.)in a volume of 5ml/kg once daily for 21days.HCE (32.5,65,130mg/kg)and the positive drug fluoxetine (15mg/kg)were administered by oral (p.o.)gavage in a volume of 10ml/kg 30min prior to the corticosterone injection for 21days.Fig.1.HPLC chromatograms of Hemerocallis citrina (A)and the standard com-pound of rutin (B).L.-T.Yi et al./Journal of Ethnopharmacology 144(2012)328–3343292.6.Sucrose preference testThe sucrose preference test was employed herein to evaluate anhedonia,one of the core symptoms of major depression in humans.The test was performed as described by Mao et al. (2012).Briefly,the sucrose preference test was carried out24h after the last drug treatment.Before the test,the rats were trained to adapt to sucrose solution(1%,w/v)by placing two bottles of sucrose solution in each cage for24h;then one of the bottles was replaced with water for24h.After the adaptation procedure,the rats were deprived of water and food for24h.The sucrose preference test was conducted at10:00a.m.The rats were housed in individual cages and given free access to the two bottles containing100ml of sucrose solution(1%,w/v)and100ml of water,respectively.After60min,the volume(ml)of both the consumed sucrose solution and water were recorded,and sucrose preference was calculated as sucrose preference(%)¼sucrose consumption(ml)/[sucrose consumption(ml)þwater consump-tion(ml)]Â100%.2.7.Forced swimming test(FST)The FST used was the same as described in detail elsewhere (Porsolt et al.,1978),with some modification.Briefly,the test was done by placing a rat in a glass cylinder(46cm in height,20cm in diameter)filled with30cm high water(25721C).Water was replaced between every trial.Two swimming sessions were conducted:an initial15min pretest,followed by a5min test 24h later.A rat was considered immobile whenever it remained floating passively in the water and only making slight movements to keep its head above the water line.The test sessions were recorded by a video camera and scored by an observer blind to treatment.2.8.Blood sampling and tissue extractionAnimals were killed by decapitation one day after FST.To avoidfluctuations on hormone levels due to circadian rhythm, animals were bled at12:00–13:00p.m.on the day of sacrifice.The brain regions of frontal cortex and hippocampus were isolated immediately,and then stored atÀ801C for later analysis of BDNF and TrkB mRNA and protein levels.2.9.Serum corticosterone assayBlood was collected on ice and separated in a refrigerated centrifuge at41C(4000Âg for10min).Serum was stored at–201C until assays were performed.Serum corticosterone levels were measured using a commercial kit(Enzo Life Sciences, Plymouth Meeting,USA)based on enzyme immunoassay (Crowther,1995).2.10.RT-PCRThe procedure of RT-PCR was performed as described in detail elsewhere(Prediger,2001;Pan et al.,2010),with some modifica-tions.Total RNA was isolated from frontal cortex or hippocampus using Trizol reagent following the manufacturer’s instructions. The concentration and purity of RNA were measured by the optical density at260and280nm using spectrophotometer. Reverse transcription was performed with1m g RNA using M-MLV reverse transcriptase for cDNA synthesis.Amplification of cDNA by PCR was performed in25m l reactions containing8m l cDNA,1m l forward and1m l reverse primers(10m m), 2.5m l PCRÂ10buffer containing MgCl2,0.5m l dNTP mixture(10mM), 0.5m l Taq polymerase(2.5U)and11.5m l sterile ddH2O.Primers used for BDNF was:50-TGTGACAGTATTAGCGAGTGGGT-30and 50-CGATTGGGTAGTTCGGCATT-30,that used for TrkB was 50-CTTATGCTTGCTGGTCTTGG-30and50-GGGTATTCTTGCTGCTC-TCA-30.In general,PCR was performed with a preheating cycle at941C for5min,denaturation,annealing and elongation were carried out at941C for30s,at501C for30s(BDNF)or591C for 60s(TrkB),and at681C for15s(BDNF)or721C for35s(TrkB), respectively.The reactions were repeated for40(BDNF)or38 (TrkB)cycles.The PCR products were resolved by a1.5%agarose gel electrophoresis and quantified by the Bio-Rad ChemiDoc XRS Gel Documentation system and Bio-Rad Quantity One software (Hercules,USA).The results were normalized to the mRNA expression level of GAPDH in each sample.2.11.Western blotThe procedure of western blot was performed as described in detail elsewhere(Blancher and Jones,2001;Chen et al.,2010), with some modifications.Brain samples were homogenized in a lysis buffer containing50mM Tris–HCl(pH7.4),1mM EDTA, 150mM NaCl,1%Triton X-100,1%sodium deoxycholate,0.1% SDS,1mM trichostatin A,phosphatase inhibitor cocktail.The homogenates were centrifuged at14000Âg for20min at41C, and the supernatants were collected and stored atÀ201C until further use.The protein concentration was determined by a BCA assay.Total proteins were separated by SDS-PAGE and transferred to a PVDF membrane.Following blocking in3%BSA/TBST at room temperature for1h,the membranes were incubated with the appropriate primary antibodies at41C overnight(anti-BDNF: 1:500,anti-TrkB:1:1000).After being washed with TBST for three times,the membranes were incubated with an HRP-labeled secondary antibody(1:4000).The blots were washed again for three times by TBST buffer and the immunoreactive bands were detected by using the enhanced chemiluminescence method. Western blot bands were scanned and subsequently analyzed densitometrically with Bio-Rad Quantity One software(Hercules, USA).The results were normalized to the protein expression level of GAPDH in each sample.2.12.Statistical analysesAll data were expressed as mean7S.E.M.To compare experi-mental and control groups,we used one-way ANOVA,followed by post-hoc Dunnett’s test.A value of P o0.05was considered statistically significant for analysis.3.Results3.1.Effects of HCE on the sucrose preferenceThe effects of HCE on the sucrose preference in rats exposed to repeated corticosterone administration were given in Fig.2. Corticosterone administration induced a significant decrease in the sucrose preference(P o0.05).65and130mg/kg HCE reversed the reduction to the baseline value(P o0.05,P o0.01,respec-tively).Fluoxetine,the positive drug,also increased the sucrose preference in the rat exposed to corticosterone(P o0.05).3.2.Effects of HCE on the immobility timeRepeated corticosterone treatment(s.c.)caused a significant elevation of immobility time in the rat FST compared with the controls(P o0.05).Long-term treatment with HCE(65and 130mg/kg)andfluoxetine(15mg/kg)for21days significantlyL.-T.Yi et al./Journal of Ethnopharmacology144(2012)328–334 330reduced the immobility time (P o 0.01,P o 0.01,P o 0.01,respec-tively)to baseline value (Fig.3).3.3.Effects of HCE on the serum corticosterone levelsExposure to corticosterone resulted in a significant increase in the serum corticosterone levels compared with the controls (P o 0.01).Long-term HCE (65and 130mg/kg)or fluoxetine (15mg/kg)treatments normalized the increased levels of corti-costerone (P o 0.05,P o 0.01,P o 0.01,respectively)in the rat exposed to repeated corticosterone administration (Fig.4).3.4.Effects of HCE on the BDNF and TrkB receptor expression To investigate the effects of HCE on the mRNA expression of BDNF and TrkB,which are two molecules crucial for antidepressant-like behavior and neurogenesis,we conducted RT-PCR and western blot analyses after the corticosterone admin-istration for 21days.As can be seen in Fig.5and 6exogenous corticosterone reduced the BDNF but not TrkB mRNA levels in the frontal cortex and hippocampus (P o 0.01,P o 0.05,respectively).HCE (65and 130mg/kg)treatments reversed the changes in the frontal cortex (P o 0.01,P o 0.05,respectively)and hippocampus (P o 0.05,P o 0.001,respectively).In addition,BDNF and TrkB protein levels in the frontal cortex (P o 0.05,P o 0.05,respectively)and hippocampus (P o 0.05,P o 0.05,respectively)were dramatically decreased by repeated corticosterone administration.HCE treatment (65,130mg/kg)alleviated the changes in the two brain regions.4.DiscussionIn general,the stress models of depression such as TST,FST and chronic mild stress are widely adopted by researchers to evaluate the antidepressant agents and their related mechanism (Willner,1997;Bai et al.,2001;Cryan et al.,2005).However,one of the biggest problems with experimenter-applied stress models is a lack of control over individual differences in responsivity to physical and psychological stressors (Sterner and Kalynchuk,2010).In such case,a more valid depression-like model needs to be established.It is well-known that elevated levels of glucocorticoid have been implicated in depression (Zunszain et al.,2011).In general,glucocorticoid serves an adaptive function in protecting the organism against many defense reactions elicited by acute stress.Once released,glucocorticoids act on bodily tissues to limit non-essential functions and mobilize energy to deal with the stressor.Glucocorticoids also reach the brain where they exert an inhibi-tory influence to halt HPA axis activity through a negative feed-back inhibitory regulation (Sterner and Kalynchuk,2010).But if the prolonged stress exists,chronically elevated glucocorticoid levels will occur and impair the negative feedback inhibition,finally lead to dysregulated emotional and physiological home-ostasis which is seen in the etiology of depression (Herman and Cullinan,1997).A number of experimental studies have also demonstrated the depression-like effects of the exogenous admin-istration of glucocorticoids.It has been observed,for instance,that corticosterone treatment (s.c.)induced depression-like behaviors in the FST and TST (Gourley et al.,2008;Lau et al.,2011).Long-term administration of corticosterone also resulted in depressed phenotype,reduced hippocampal cell proliferation and granule cell layer volume,as observed in patients suffering from depression (Murray et al.,2008).Thus,the findings suggest that administra-tion of exogenous corticosterone is a useful method for studying the relationship between stress,behaviors,glucocorticoids,neuro-trophic factors and depression (Sterner and Kalynchuk,2010).In our present study,repeated corticosterone treatment induced a reduction of sucrose preference and an increase of immobility time,which is in line with other findings (Gourley et al.,2008;Mao et al.,2012).The sucrose preference,which reflects a symptom anhedonia (inability to experience pleasure)existing in depressed patients,can be improved by antidepres-sants (Willner,1997).The present findings revealed that repeated administration of HCE reversed the anhedonic-like behavior in corticosterone-induced rats.In the other behavioral test,the measurement index immobility time of the FST,which isinducedFig.3.Effects of HCE on the immobility time in the rat FST.#P o 0.05vs control with vehicle group.**P o 0.01vs corticosterone with vehiclegroup.Fig.4.Effects of HCE on the serum corticosterone levels (one day after FST)in the rats exposed to repeated corticosterone administration.##P o 0.01vs control with vehicle group.*P o 0.05and **P o 0.01vs corticosterone with vehiclegroup.Fig.2.Effects of HCE on the sucrose preference in the rats exposed to repeated corticosterone administration.#P o 0.05vs control with vehicle group.*P o 0.05and **P o 0.01vs corticosterone with vehicle group.L.-T.Yi et al./Journal of Ethnopharmacology 144(2012)328–334331by swimming stress,can be dramatically decreased by a broad spectrum of clinically effective antidepressants (Hasco¨et and Bourin,2009).Consistent with our previous study (Gu et al.,2012),HCE also reversed the elevation of immobility time induced by corticosterone treatment (s.c.)in the rat FST.Taken together,the results obtained from the behavioral studies again confirmed that HCE treatment exerted an antidepressant-like effect in the rats treated by corticosterone (s.c.).It was observed in our study that repeated corticosterone treatment (s.c.)caused an elevation of serum corticosterone levels in rats,which was consistent with the previous studies (Dwivedi et al.,2006;Johnson et al.,2006).This means that exogenous corticosterone results in an absolute increase in circulating serum corticosterone levels,an indicator of stress and depression in laboratory animals.Thus,the present study revealed that the behavioral consequences of repeated corticosterone administra-tion were accompanied by dysregulation of the HPA axis.Glucocorticoid and glucocorticoid receptor (GR)has been considered to be especially important in relation to BDNF regula-tion (Numakawa et al.,2009).Stressors stimulate the activity of the HPA axis,and then glucocorticoids increase,which lead to the reduced expression of BDNF (Lee and Kim,2010).For example,previous study indicated drinking corticosterone decreased brain BDNF expression (Gourley et al.,2008).It was also observed that adrenalectomy resulted in an elevation in the levels of BDNF (Chao et al.,1998).Consistent with the studies,our study also found that administration of exogenous corticosterone reduced the BDNF mRNA and protein levels in the frontal cortex and hippocampus of rats.In addition,increases of BDNF in brain have been shown to attenuate depression-related behavior (Shirayama et al.,2002).Moreover,antidepressant increases BDNF-induced hippocampal neurogenesis via a GR-dependent mechanism (Anacker et al.,2011).Therefore,when we examined the effects of HCE on the corticosterone-mediated decrease in BDNF expres-sion,the results of long-term HCE treatment reversing the reduction suggested that the behavioral improvement in the model may be concurrent with increased BDNF levels.Growing evidence suggests BDNF-TrkB signaling serves an important role in the regulation of many of the behavioral and molecular mechanism of antidepressant (Saarelainen et al.,2003).Either decreased BDNF availability or reduced levels of TrkB neurotrophin receptor could reduce BDNF signaling (Fahnestock et al.,2012).Short-or long-term treatments with antidepressants activate TrkB receptor in mouse brain (Rantam¨aki et al.,2007).Therefore,besides BDNF,we evaluated the effects of HCE on gene and protein expressions of TrkB receptor in the frontal cortex and hippocampus.The immunoblotting results showed that TrkB receptor proteins in frontal cortex and hippocampus were mark-edly decreased by repeated corticosterone administration and reversed by HCE treatment.However,there were no significant changes in frontal cortex and hippocampal TrkB mRNA levels after long-term corticosterone or HCE treatment.The results0.00.51.01.5HCEVehicle Vehicle1532.565130FluoxetineControlCorticosterone administration R e l a t i v e B D N F m R N A l e v e l s (n o r m a l i z e d b y G A P D H )0.00.51.01.5(mg/Kg)(mg/Kg)(mg/Kg)HCEVehicle Vehicle1532.565130FluoxetineControlCorticosterone administrationR e l a t i v e B D N F p r o t e i n l e v e l s (n o r m a l i z e d b y G A P D H )0.00.51.01.5HCEVehicle Vehicle1532.565130FluoxetineControlCorticosterone administration R e l a t i v e T r k B m R N A l e v e l s (n o r m a l i z e d b y G A P D H )(mg/Kg)HCEVehicleVehicle1532.565130FluoxetineControlCorticosterone administrationR e l a t i v e T r k B p r o t e i n l e v e l s (n o r m a l i z e d b y G A P D H)Fig.5.Effects of HCE on the frontal cortex BDNF and TrkB receptor mRNA and protein expressions in the rats exposed to repeated corticosterone administration.#P o 0.05and ##P o 0.01vs control with vehicle group.*P o 0.05,**P o 0.01and ***P o 0.001vs corticosterone with vehicle group.L.-T.Yi et al./Journal of Ethnopharmacology 144(2012)328–334332indicated that long-term corticosterone affected the rate of TrkB protein synthesis or degradation but not TrkB mRNA,which was partly consistent with the previous study by Kutiyanawalla et al.(2011).As a result,these data suggest HCE-induced antidepres-sant-like effect is likely through modulation of the BDNF-TrkB neurotrophic signaling pathway.However,the related mechan-ism involved in the regulation of TrkB mRNA and protein should be performed in further studies.5.ConclusionTaken together,the data from our study indicate that repeated corticosterone administration induces a decrease in BDNF and TrkB expressions as well as depression-related behaviors in rats.More-over,the behavioral changes attenuated by long-term HCE treat-ment are possibly mediated by increasing BDNF-TrkB signaling in the frontal cortex and hippocampus.Further studies will be neces-sary to improve our understanding of the correlation between HCE treatment and the BDNF induced hippocampal neurogenesis.AcknowledgmentThe project was supported by the Natural Science Foundation of Fujian Province of China (No.2011J05081)and HuaqiaoUniversity (No.09BS507).And we thank Dr.Di Geng and Mr.Hao Cheng for technical assistance in HPLC analysis of Hemerocallis citrina .ReferencesAnacker,C.,Zunszain,P.A.,Cattaneo,A.,Carvalho,L.A.,Garabedian,M.J.,Thuret,S.,Price,J.,Pariante,C.M.,2011.Antidepressants increase human hippocampal neurogenesis by activating the glucocorticoid receptor.Molecular Psychiatry 16,738–750.Bai, F.,Li,X.,Clay,M.,Lindstrom,T.,Skolnick,P.,2001.Intra and interstraindifferences in models of ‘‘behavioral despair’’.Pharmacology Biochemistry and Behavior 70,187–192.Blancher,C.,Jones,A.,2001.SDS-PAGE and western blotting techniques.Methodsin Molecular Medicine 57,145–162.Chao,H.M.,Sakai,R.R.,Ma,L.Y.,McEwen,B.S.,1998.Adrenal steroid regulation ofneurotrophic factor expression in the rat hippocampus.Endocrinology 139,3112–3118.Chen,C.C.,Hsu,L.W.,Huang,L.T.,Huang,T.L.,2010.Chronic administration ofcyclosporine A changes expression of BDNF and TrkB in rat hippocampus and midbrain.Neurochemical Research 35,1098–1104.Chen,G.Z.,Zhang, C.X.,Fan,Y.S.,Chen,J.B.,2008.Treatment of 156cases ofdepression related insomnia by Hemerocallis citrina .Zhejiang Journal of Traditional Chinese Medicine 43,397.Crowther,J.R.,1995.ELISA.Theory and practice.Methods in Molecular Biology 42,1–218.Cryan,J.F.,Mombereau,C.,Vassout,A.,2005.The tail suspension test as a modelfor assessing antidepressant activity:review of pharmacological and genetic studies in mice.Neuroscience and Biobehavioral Reviews 29,571–625.Dwivedi,Y.,Rizavi,H.S.,Pandey,G.N.,2006.Antidepressants reversecorticosterone-mediated decrease in brain-derived neurotrophic factor0.00.51.01.5(mg/Kg)HCEVehicle Vehicle1532.565130FluoxetineControlR e l a t i v e B D N F m R N A l e v e l s (n o r m a l i z e d b y G A P D H )0.00.51.01.5(mg/Kg)HCEV e hicl e V e hicl e15 32.5 65 130FluoxetineControlR e l a t i v e B D N F p r o t e i n l e v e l s (n o r m a l i z e d b y G A P D H )0.00.51.01.5(mg/Kg)HCEVehicle Vehicle1532.565 130FluoxetineControlR e l a t i v e T r k B m R N A l e v e l s (n o r m a l i z e d b y G A P D H )0.00.51.01.5(mg/Kg)HCEVehicle Vehicle1532.565130FluoxetineControlR e l a t i v e T r k B p r o t e i n l e v e l s(n o r m a l i z e d b y G A P D H)Corticosterone administration Corticosterone administrationCorticosterone administration Corticosterone administrationFig.6.Effects of HCE on the hippocampal BDNF and TrkB receptor mRNA and protein expressions in the rats exposed to repeated corticosterone administration.#P o 0.05vs control with vehicle group.*P o 0.05,**P o 0.01and ***P o 0.001vs corticosterone with vehicle group.L.-T.Yi et al./Journal of Ethnopharmacology 144(2012)328–334333。
64曲阜师范大学2020年成人高等教育 《英语词汇学 》复习资料 期末考试试题及参考答案
《英语词汇学》复习资料1Ⅰ. Fill in the blanks.Directions: Complete the following statements with proper words.1.The 1 is the smallest functioning unit in the composition of words.2. 2 are words borrowed early in the past and now are well assimilated into the Englishlanguage.3.The problem of interrelation of the various meanings of the same word can be dealt with fromtwo different angles: 3 approach and synchronic approach.4.“Mal-” in “maltreat” is a 4 prefix, while “inter-” in “ interstate” is a 5 prefix.5.Old English is described as a language of full endings, Middle English language of 6endings, and a language of 7 endings.6.In modern English, one may find some 8 words whose sounds suggest their meaning,for these words were created by imitating the natural sounds or noises.7.The word meaning is made up of 9 meaning and 10 meaning, and the later hastwo components: conceptual meaning and 11 meaning.8.Words that have emotive values may fall into two categories: appreciative or 12 .9.13 is thought to be the opposite process of suffixation.10.14 is the formation of new words by combining parts of two words or a word plus apart of another word.11.15 refers to the jargon of criminals. Its use is confined to the sub-cultural groups, andoutsiders can hardly understand it.12.“Pretty” and “handsome” share the same 16 meaning，but differ in 17 meaning.13.___18___analysis is a process of breaking down the sense of a word into its minimalcomponents which are also known as semantic features..14.Radiation and 19 are the two coinages which the development of word meaningfollows from monosemy to polysemy.15.20 deals with the relationship of inclusion, i.e. the meaning of a more specific word isincluded in that of another more general word.Ⅱ. Decide whether the following statements are true or false. Write T for “true” and F for “false”.1.Homonyms are descendants of different sources whereas a polysemant is a word of the samesource which has acquired different meanings in the course of development.2.Words of the basic word stock are mostly root words or monosyllabic words, so they havestrong productivity.3.“Can-opener” used as slang to mean “all-purpose key”.4.Native words are neutral in style.5.The Indo-European language family is made up of most languages of Europe, the Far East, andIndia.6.Borrowing has played a vital role in the development of English vocabulary, particularly inearlier times.7.The smallest functioning unit in the composition of words is morpheme.8.Stem is a form to which affixes of any kind can be added.9.Base is what remains of a word after the removal of all affixes.10.Words created by compounding occupy the highest percentage of the English vocabulary.11.“Fore-” in “forehead” and “fore-” in “foreknowledge” belong to two kinds of prefix.12.Word-building and word-formation are relative synonyms.13.The word manusc ript which originally denotes “handwriting” only has undergone a process ofextension of meaning.14.Parent—child and husband—wife are two pairs of converses.15.Policeman, constable, bobby and cop are synonyms differing in intensity.Ⅲ. Answer the following questions briefly.1.What are the characteristics of the basic word stock?2.Why are prefixes and suffixes divided according to different criteria?3.List the four sources of synonyms.4.What are the characteristics of conceptual meaning and associative meaning?Ⅳ. Answer the following questions according to the requirement.Classify the three pairs of antonyms according to types of antonyms you have learned and describe the characteristics of each type of them.interviewer/interviewee; male/female; old /young答案I．Fill in the blanks.1. morpheme2. denizens3. diachronic4. pejorative5. locative6. leveled7. lost8. onomatopoeic9. grammatical 10.lexical11.associative 12. pejorative 13. backformation 14. blending 15.argot 16. conceptual 17. collocative 18. componential 19. concatenation 20.hyponymyII.Decide whether the following statements are true or false. Write T for “true” and F for “false”.1-5 TTTFT 6-10 TFFFT 11-15 TFFTFIII.Answer the following questions briefly.1.What are the characteristics of the basic word stock?1)All national character 2) stability 3) productivity 4) polysemy 5)collocability2.Why are prefixes and suffixes divided according to different criteria?1)Prefixes primarily effect a semantic modification of the base, i.e. prefixes do not generallychange the word-class of the base but only modify its meaning.2)Suffixes have only a small semantic role and their primary function is to change thegrammatical function of the base, i.e. the change of the word class with a slight modification of meaning.3)So prefixes are categorized on a semantic basis while suffixes are divided on a grammaticalbasis.3.1)Borrowing; (2) dialects and regional English (3) figurative and euphemistic use of words(4) coincidence with idiomatic expressions4.What are the characteristics of conceptual meaning and associative meaning?1)Conceptual meaning is the meaning given in the dictionary and forms the core of wordmeaning. Being constant and relatively stable, conceptual meaning forms the basis for communication as the same word generally has the same conceptual meaning to the speakers in the same speech community. (3%)2)Associative meaning differs from the conceptual meaning in that it is open-ended andindeterminate, liable to the influence of such factors as culture, experience, religion,geographical region, class background, education, etc…（3%）Ⅳ. Analyze the following questions and explain them according to the requirement.1.1)Interviewer& interviewee are converses; male & female are complementaries; old &young are contraries.2)Complementaries truly represent oppositeness of meaning. They are so opposite to eachother that they are mutually exclusive and admit no possibility between them. The assertion of one is the denial of the other or vice versa. Complementaries are nongradable, and they cannot be used in comparative degrees and do not allow adverbs of intensity like “very” to qualify them.3)Contraries are gradable antonyms. The existence of one is in relation to the other. We cansay: A man is rich or very rich and also we can say a man is rich than the other. Contraries are characteristic of semantic polarity. These antonyms form part of a scale of values between two poles and can accommodate a middle ground belonging neither to one pole nor to the other.4)Converses consist of relational opposites. The pairs of words indicate reciprocal socialrelationships that one of them cannot be used without suggesting the other. It also includes reverse terms, which comprise adjectives and adverbs signifying a quality or verbs and nouns signifying an act or state that reverse or undo the quality, action or state of the other.复习资料2I. 单选题1. In the sentence “I like to see a movie.”, there are ________ functional words.A. 2B. 3C. 4D. 52. Conversion is amethod________________________.A. of turning words of one part of speech to those of a different part of speechB. of converting words of one meaning into different meaningC. of deriving words through grammatical meansD. of changing words in morphological structure3. The following words have derivational affixes EXCEPT ________________.A. subseaB. prewarC. postwarD. desks4. Which of the following statements is false?A. Conversion refers to the use of words of one class as that of a different class.B. Words mainly involved in conversion are nouns, verbs and adverbs.C. Partial conversion and full conversion are concerned with adjectiveswhen converted to nouns.D. The conversion between nouns and verbs may involve a change of stress.5. _________ is the meaning given in the dictionary and forms the core of word-meaning.A. Grammatical meaningB. Denotative meaningC. Associative meaningD. Connotative meaning6. The words what have emotive content in themselves are said to contain __ meaning.A. collocativeB. affectiveC. stylisticD. denotative7. __________ explains the connection between the literal sense and figurative sense of the word.A. Etymological motivationB. Onomatopoetic motivationC. Morphological motivationD. Semantic motivation8. The following words have inflectional affixes EXCEPT __________.A. worksB. workerC. workingD. worked9. “Smog”is formed by combining “smoke”and “fog”. So it is an example ofA. clippingB. compoundingC. blendingD. back-formation10. The word “smog”is created by blending, with the structure of __________.A. head + tailB. head + headC. head + wordD. word + tail11. The most important mode of vocabulary development in present-day English is the creation of new words by means of ________________.A. translation-loansB. emantic loansC. word formationD. borrowings12. Which of the following belongs to a semantic field?A. steed, charger, palfrey, plug, nagB. pony, mustang, mule, stud, mareC. policeman, constable, bobby, copD. domicile, residence, abode, home13. Words which are used to show the attitude of approval are ________________.A. appreciativeB. pejorativeC. conntativeD. collocative14. General features of English contains the following except _________.A. simplicityB. receptivityC. adaptabilityD. imprssiveness15. The most productive means of word-formation in modern English are the following except .A. compoundingB. affixationC. acronymD. conversionII判断题1. The Indo-European language family is made up of most languages of Europe, the Far East, andIndia. （）2. The word manusc ript which originally denotes “handwriting” only has undergone a process ofextension of meaning. （）3. The beginning of the Middle English Period was marked by the Norman Conquest which broughtmany Latin words into the English language. （）4. Words of the basic word stock are mostly root words or monosyllabic words, so they have strongproductivity. （）5. Grammatical meaning or a word includes part of speech, tense meaning, and stylistic coloring.（）6. Words created by compounding occupy the highest percentage of the English vocabulary. （）7. The marked term of each pair of antonyms covers the sense of the unmarked term. （）8. Policeman, constable, bobby and cop are synonyms differing in intensity. （）9. Borrowing has played a vital role in the development of English vocabulary, particularly inearlier times. （）10. “Radiation” shows that the derived mea nings of a polysemantic word are not directly related tothe primary meaning. （）III简答题1. What are the characteristics of conceptual meaning and associative meaning?2. List different types of associative meaning and define them.答案I． 1-5 AADDB 6-10 BDBCA 11-15 CBADCⅡ. 1-5 TFFTF 6-10 TFFTFⅢ. 1. What are the characteristics of conceptual meaning and associative meaning?Conceptual meaning is the meaning given in the dictionary and forms the core of word meaning.Being constant and relatively stable, conceptual meaning forms the basis for communication as the same word generally has the same conceptual meaning to the speakers in the same speech community. Associativemeaning differs from the conceptual meaning in that it is open-ended and indeterminate, liable to the influence of such factors as culture, experience, religion, geographical region, class background, education, etc…2. List different types of associative meaning and define them.Explain different types of homonyms with examples.Perfect homonyms are known as absolute homonyms, and they are words identical both in sound and spelling. E.g bear (to put up with) and bear (a kind of fruit) Homographs are words identical only in spelling but different in sound and meaning, e.g. sow (to scatter seeds) and sow (female adult pig)Homophones are words identical only in sound but different in spelling and meaning, e.g. dear ( a loved person) and deer (a kind of an animal)复习资料3I．Fill in the blanks.Directions: Complete the following statements with proper words.1.The __1 is the smallest functioning unit in the composition of words.2. 2 are words borrowed early in the past and now are well assimilated into the Englishlanguage.3.The problem of interrelation of the various meanings of the same word can be dealt withfrom two different angles: 3 approach and synchronic approach.4.“Mal” in “maltreat” is a 4 prefix, while “inter-” in “ interstate” is a 5_ prefix.5.Old English is described as a language of full endings, Middle English language of___6__endings, and a language of __7__ endings.6.In modern English, one may find some 8 words whose sounds suggest their meaning,for these words were created by imitating the natural sounds or noises.7.The word meaning is made up of 9 meaning and 10 meaning, and the later hastwo components: conceptual meaning and 11 meaning.8.Words that have emotive values may fall into two categories: appreciative or __12 .9.13 is thought to be the opposite process of suffixation.10.___14__ is the formation of new words by combining parts of two words or a word plus apart of another word.11.15 refers to the jargon of criminals. Its use is confined to the sub-cultural groups, andoutsiders can hardly understand it.12.“Pretty” and “handsome” share the same _16_ meaning, but differ in _17_ meaning.13.___18___analysis is a process of breaking down the sense of a word into its minimalcomponents which are also known as semantic features.14.Radiation and ___19___ are the two coinages which the development of word meaningfollows from monosemy to polysemy.15.__20____deals with the relationship of inclusion, i.e. the meaning of a more specific word isincluded in that of another more general word.Ⅱ. Decide whether the following statements are true or false and write T or F on the answer sheet:1.Homonyms come mainly from borrowing, changes in sound and spelling, and dialects.2.“Radiation” shows that the derived meanings of a polysemantic word are not directly related tothe primary meaning.3.Borrowing is a very important source of synonyms.4. A word which has a synonym naturally has an antonym.5.Hyponymy deals with the relationship of semantic inclusion.6.Motivation explains the connection between the linguistic form and its meaning.7.Grammatical meaning or a word includes part of speech, tense meaning, and stylistic coloring.8.The origins of the words are a key factor in distinguishing homonyms from polysemants.9.The marked term of each pair of antonyms covers the sense of the unmarked term.10.If the words differ in range and intensity of meaning, the words are not identical in denotation.11.The beginning of the Middle English Period was marked by the Norman Conquest whichbrought many Latin words into the English language.ponential analysis is to break down. the conceptual sense of a word into its minimaldistinctive components.13.Celtic language made great contributions to the expansion of the English vocabulary.14.Native words enjoy the same features as the basic word stock and more.15.Shortening includes clipping and blending.Ⅲ. Answer the following questions briefly.1. Analyze the morphological structures of the following words and point out the types of the morphemes in terms of free and bound morphemes.unbearable international ex-prisoner.2. How would you explain the difference between back formation and suffixation? Give examples to illustrate your point.3. List different types of associative meaning and define them.4. Explain different types of homonyms with examples.Ⅳ. Analyze the following questions and explain them according to the requirement.1. What is the difference between homonyms and polysemants?答案I．Fill in the blanks.1. morpheme2. denizens3. diachronic4. pejorative5. locative6. leveled7. lost8. onomatopoeic9. grammatical 10. lexical 11.associative 12. pejorative 13. backformation 14. blending 15. argot 16. conceptual 17. collocative 18. componential 19. concatenation 20. hyponymyⅡ. Decide whether the following statements are true or false and write T or F in the brackets: 1．F 2.F 3.T 4.F 5.T 6. T 7.F 8.T 9.F 10.T 11.F 12. F 13. F 14. T 15. TⅢ. Answer the following questions briefly.1. Analyze the morphological structures of the following words and point out the types of the morphemes in terms of free and bound morphemes.unbearable international ex-prisoner.un+bear+able:(1)‘bear’ is a free morpheme, and ‘un’, ‘able’are bound morphemes.inter+nation+al: ‘nation’ is a free morpheme, and ‘inter, al’ are bound morphemes.ex+prison+er: ‘prison’ is a free morpheme, and ‘ex, er’ are bound morphemes.2. How would you explain the difference between back formation and suffixation? Give examples to illustrate your point.1)Back-formation is considered to be the opposite process of suffixation.2)Suffixation is the formation of new words by adding suffixes to bases.3)Backformation is therefore the method of creating words by removing the supposed suffixes, socalled because many of the removed endings are not suffixes but inseparable parts of the word.4)For example, it is a common practice to add –er, -or to verb bases to form agential nouns.5)Reasonably, people make verbs by dropping the ending such as –or in editor, -ar in beggar and–er in butler.3. List different types of associative meaning and define them.1)Connotative meaning refers to the overtones or associations suggested by the conceptualmeaning, traditionally known as connotations.2)Stylistic meaning refers to stylistic features, which make them appropriate for different styles.3)Affective meaning expresses the speaker’s attitude towards the person or thing in question.4)Collocative meaning consists of the associations a word acquires on account of the meanings ofwords which tend to occur in its environment.4. Explain different types of homonyms with examples.(1)Perfect homonyms are known as absolute homonyms, and they are words identical both insound and spelling. E.g bear (to put up with) and bear (a kind of fruit)(2)Homographs are words identical only in spelling but different in sound and meaning, e.g. sow(to scatter seeds) and sow (female adult pig)(3)Homophones are words identical only in sound but different in spelling and meaning, e.g. dear( a loved person) and deer (a kind of animal)Ⅳ. Analyze the following questions and explain them according to the requirement.1.What is the difference between homonyms and polysemants?1)Perfect homonyms and polysemants are fully identical with reference to spelling andpronunciation, as both have the same orthographical form but different meanings. This creates the problem of differentiation.2)The fundamental difference between homonyms and polysemants lies in the fact that the formerrefers to different lexemes which have the same form and the latter the one and same lexeme which has several distinguishable meanings.3)One important criterion by which to differentiate them is ‘etymology’, i.e., homonyms aredescendants of different sources whereas a polysemant is a word of the same source which has acquired different meanings in the course of development.4)The second principal consideration is ‘semantic relatedness’. The several meanings of a singlepolysemous lexeme are related and can be traced back to one central meaning. On the other hand, meanings of different homonyms have nothing to do with one another.5)In dictionaries, a polysemant has its meanings all listed under one headword whereashomonyms are listed as separate entries.。
Python中的自然语言处理技术及应用实例
Python中的自然语言处理技术及应用实例自然语言处理（Natural Language Processing，NLP）是人工智能领域的一个重要分支，它致力于让计算机能够理解、处理和生成人类语言。
Python作为一种功能强大且易于使用的编程语言，在自然语言处理领域得到广泛应用。
本文将介绍Python中的自然语言处理技术，并提供应用实例。
一、文本预处理在进行自然语言处理之前，对文本进行预处理是必要的。
Python中有丰富的库和工具可用于文本预处理，如NLTK（Natural Language Toolkit）和spaCy。
下面以NLTK为例，介绍几个常用的文本预处理技术。
1. 分词（Tokenization）分词是将文本分解成一个个有意义的单词或符号的过程。
NLTK提供了多种分词方法，如基于规则的分词、基于正则表达式的分词以及基于机器学习的分词。
以下是一个使用NLTK进行分词的实例：```pythonimport nltktext = "This is a sample sentence."tokens = nltk.word_tokenize(text)print(tokens)```输出结果为：['This', 'is', 'a', 'sample', 'sentence', '.']2. 去除停用词（Stopword Removal）停用词是指在文本中频繁出现，但往往没有实际含义的词，如"the"、"and"等。
在进行自然语言处理时，通常需要去除这些停用词，以减少干扰。
NLTK提供了一份常用的英文停用词列表，可以根据需要进行自定义。
以下是一个使用NLTK去除停用词的实例：```pythonfrom nltk.corpus import stopwordsstop_words = set(stopwords.words('english'))filtered_tokens = [word for word in tokens if word.lower() not instop_words]print(filtered_tokens)```输出结果为：['sample', 'sentence', '.']3. 词干提取（Stemming）词干提取是将一个单词的不同变体转化为其原始形式（即词干）的过程。
学术英语(第二版)医学教师用书Unit 8
UNIT 8 Principles of Biomedical EthicsTeaching ObjectivesAfter learning Unit 8, Ss are expected to accomplish the following objectives:To know the boundaries between medical research and practice To have a clear understanding of the moral principles and behavioral guidelines for the biomedical research and medical practice To understand the boundaries drawn between medical research and practice To know three basic ethical principles of research involving human subjects To learn the requirements when basic principles are properly applied in research To be more prepared for a life-or-death decision in medical practice To get more insights into the ethical justification of dilemmas in medical practice To know some building blocks in medical terminologyTo be familiar with expressions used to define key termsTo further develop awareness of formal and informal language To get familiar with the Cornell note-taking system To know the two approaches to medical decisions: traditional paternalistic mode and more recent collaborative modeTo learn how to develop a strong conclusion To know the format requirements of the reference listTo be able to make a reference list according to style requirements To be aware of the balance between medical authority and patients ’ autonomyProfessionalknowledgeReadingAcademic vocabulary anddiscourseViewingSpeakingWritingResearchingTeaching Activities and ResourcesPart 1 ReadingText ALead-inSuggested teaching plan1. To draw Ss’ attention and to raise their awareness of the importance ofbiomedical ethics, T is advised to relate the discussion of this unit to the real-world happenings.Before starting the class,search the media for the latest news reports,either at home or abroad,about controversial events in medicine community or healthcare settings.2. Start the class by doing Task / Lead-in and relate the content of the video clip toyour findings in the pre-class searching.Key to the task2) Death4) Patient rightsScriptWell,advancements in medical science have afforded us the opportunity to live decades longer than in previous generations.For every new possibility offered, we now face an equal number of challenges and we find ourselves confronting decisions that are unprecedented in human history.When does life begin?When should life end? How do we define death when we have the ability to keep people technically alive,or we should say,technologically alive long after their discrete body parts no longer function? Welcome to “Matter and Beyond . ” I’m your host MaryLynn Schiavi.In this program we’re going to explore issues around medical science that are forcing us to define life, death, quality of life, patient rights, and confront the moral and ethical questions that arise when facing critical healthcare decisions.3. Introduce the topic of Text A as a natural continuum of Lead-in .Text Comprehension1. Make good use of Lead-in video clip as it serves as a perfect introduction to thetopic of this unit. Elaborate on the connection of its content with the latest events in the real world. Naturally, ask Ss how medicine differs from other branches of natural science, especially when human subjects are involved in the research. Here are some hints:2. Analyze the text and lead Ss to discuss, integrating Task 2 / Critical reading andthinking / Text A into analysis and discussion. The presentation topics should be assigned to individual Ss for preparation at least one week in advance. Ask other Ss to preview the text with the guidance of presentation topics.3. Integrate Task 2 / Language building-up / Text A when a careful definition ofkey terms is covered.4. When analyzing the text, ask Ss to pay special attention to the sentences listed inLanguage focus below.5. If time allows, ask Ss to do Task 1 / Critical reading and thinking / Text A inabout five minutes. Check out the task by asking one or two Ss to read their answers. This is done to get an overview about the text.Language focus 1. … described in a formal protocol that sets forth an objective … (P185, Para.2)set forth 是动词词组，表示用清晰、具体的方式解释或描述，多用于正式的 书面语中。
英语语言学名词解释补充
Chapter 11 : Second Language Acquisition1. second language acquisition:It refers to the systematic study of how one person acquires a second language subsequent to his native language.2. target language: The language to be acquired by the second language learner.3. second language:A second language is a language which is not a native language in a country but which is widely used as a medium of communication and which is usually used alongside another language or languages.4. foreign language:A foreign language is a language which is taught asa school subject but which is not used as a medium of instruction in schools nor as a language of communication within a country.5. interlanguage: A type of language produced by second and foreign language learners, who are in the process of learning a language, and this type of language usually contains wrong expressions.6. fossilization: In second or foreign language learning, there is a process which sometimes occurs in which incorrect linguistic features become a permanent part of the way a person speaks or writes a language.7. contrastive analysis: a method of analyzing languages for instructional purposes whereby a native language and target language are compared with a view to establishing points of difference likely to cause difficulties for learners.8. contrastive analysis hypothesis: A hypothesis in second language acquisition. It predicts that where there are similarities between the first and second languages, the learner will acquire second language structure with ease, where there are differences, the learner will have difficulty.9. positive transfer:It refers to the transfer that occur when both the native language and the target language have the same form, thus making learning easier. (06F)10. negative transfer:the mistaken transfer of features of one’s native language into a second language.11. error analysis: the study and analysis of errors made by second and foreign language learners in order to identify causes of errors or common difficulties in language learning.12. interlingual error:errors, which mainly result from cross-linguistic interference at different levels such as phonological, lexical, grammatical etc.13. intralingual error:Errors, which mainly result from faulty or partial learning of the target language, independent of the native language. The typical examples are overgeneralization and cross-association.14. overgeneralization:The use of previously available strategies in new situations, in which they are unacceptable.15. cross-association: some words are similar in meaning as well as spelling and pronunciation. This internal interference is called cross-association.16. error: the production of incorrect forms in speech or writing by a non-native speaker of a second language, due to his incomplete knowledge of the rules of that target language.17. mistake:mistakes, defined as either intentionally or unintentionally deviant forms and self-corrigible, suggest failure in performance.18. input: language which a learner hears or receives and from which he or she can learn.19. intake: the input which is actually helpful for the learner.20. Input Hypothesis:A hypothesis proposed by Krashen , which states that in second language learning, it’s necessary for the learner to understand input language which contains linguistic items that are slightly beyond the learner’s present linguistic competence. Eventually the ability to produce language is said to emerge naturally without being taught directly.21. acquisition: Acquisition is a process similar to the way children acquire their first language. It is a subconscious process without minute learning of grammatical rules. Learners are hardly aware of their learning but they are using language to communicate. It is also called implicit learning, informal learning or natural learning.22. learning: learning is a conscious learning of second languageknowledge by learning the rules and talking about the rules.23. comprehensible input:Input language which contains linguistic itemsthat are slightly beyond the learner’s present linguistic competence.(06F)24. language aptitude: the natural ability to learn a language, notincluding intelligence, motivation, interest, etc.25. motivation:motivation is defined as the learner’s attitudes andaffective state or learning drive.26. instrumental motivation: the motivation that people learn a foreignlanguage for instrumental goals such as passing exams, or furthering acareer etc. (06C)27. integrative motivation:the drive that people learn a foreign languagebecause of the wish to identify with the target culture. (06C/ 05)28. resultative motivation: the drive that learners learn a secondlanguage for external purposes. (06F)29. intrinsic motivation: the drive that learners learn the secondlanguage for enjoyment or pleasure from learning.30. learning strategies:learning strategies are learners’ consciousgoal-oriented and problem-solving based efforts to achieve learningefficiency.31. cognitive strategies: strategies involved in analyzing, synthesis,and internalizing what has been learned. (07C/ 06F)32. metacognitive strategies:the techniques in planning, monitoring andev aluating one’s learning.33. affect/ social strategies: the strategies dealing with the wayslearners interact or communicate with other speakers, native ornon-native.Chapter 12 : Language And Brain1. neurolinguistics: It is the study of relationship between brain andlanguage. It includes research into how the structure of the braininfluences language learning, how and in which parts of the brain language is stored, and how damage to the brain affects the ability to use language.2. psycholinguistics: the study of language processing. It is concerned with the processes of language acqisition, comprehension and production.3. brain lateralization: The localization of cognitive and perceptive functions in a particular hemisphere of the brain.4. dichotic listening:A technique in which stimuli either linguistic or non-linguistic are presented through headphones to the left and right ear to determine the lateralization of cognitive function.5. right ear advantage: The phenomenon that the right ear shows an advantage for the perception of linguistic signals id known as the right ear advantage.6. split brain studies: The experiments that investigate the effects of surgically severing the corpus callosum on cognition are called as split brain studies.7. aphasia: It refers to a number of acquired language disorders due to the cerebral lesions caused by a tumor, an accident and so on.8. non-fluent aphasia:Damage to parts of the brain in front of the central sulcus is called non-fluent aphasia.9. fluent aphasia: Damage to parts of the left cortex behind the central sulcus results in a type of aphasia called fluent aphasia.10. Acquired dyslexia: Damage in and around the angular gyrus of the parietal lobe often causes the impairment of reading and writing ability, which is referred to as acquired dyslexia.11. phonological dyslexia:it is a type of acquired dyslexia in which the patient seems to have lost the ability to use spelling-to-sound rules.12. surface dyslexia: it is a type of acquired dyslexia in which the patient seems unable to recognize words as whole but must process all words through a set of spelling-to-sound rules.13. spoonerism:a slip of tongue in which the position of sounds, syllables, or words is reversed, for example, Let’s have chish and fips instend of Let’s have fish and chips.14. priming: the process that before the participants make a decision whether the string of letters is a word or not, they are presented with an activated word.15. frequency effect: Subjects take less time to make judgement on frequently used words than to judge less commonly used words . This phenomenon is called frequency effect.16. lexical decision: an experiment that let participants judge whethera string of letter is a word or not at a certain time.17. the priming experiment:An experiment that let subjects judge whethera string of letters is a word or not after showed with a stimulus word, called prime.18. priming effect:Since the mental representation is activated through the prime, when the target is presented, response time is shorter that it otherwise would have been. This is called the priming effect. (06F)19. bottom-up processing: an approach that makes use principally of information which is already present in the data.20. top-down processing:an approach that makes use of previous knowledge and experience of the readers in analyzing and processing information which is received.21. garden path sentences: a sentence in which the comprehender assumesa particular meaning of a word or phrase but discovers later that the assumption was incorrect, forcing the comprehender to backtrack and reinterpret the sentence.22. slip of the tongue:mistakes in speech which provide psycholinguistic evidence for the way we formulate words and phrases.。
肖仰华_大数据时代的人工智能与智能医疗
大数据人工智能的机遇中国医疗现状的问题智能医疗的实现路径大数据人工智能技术的挑战问题信任均衡多因素医生病人医院社会精准医疗面临的挑战足够的资源病理学家的匮乏超负荷工作大量的各类数据我们正在经历！终极目标将答案交给机器：海量数据作为学习样本，不断优化模型贴近答案代表：基于深度学习的系统将目标交给机器：机器拥有自我学习，自我完善的能力，写出能够完善自身的代码，不断从开放世界获取信息并改造自身。
早期人工智能将规则交给机器：给定规则，机器根据规则作出相应判断代表：规则系统和专家系统完整生态深度学习模型：具有较深层次的神经网络模拟人脑的深度结构;模仿人类逐层进行，逐步抽象的认知过程使得类人学习成为可能无需传统机器学习的专家特征定义是一种大规模语义网络，表达了实体/概念及其之间的各类语义关系海量规模、语义丰富、结构友好为语义理解提供丰富的背景知识是能够满足机器认知需求的知识表示路径1.Understands natural language and human communicationcognitive services •Ask 智慧搜索智能推荐精准画像自然交互论坛十亿级文献一亿级病例百万级自然语言文本杂志/论文/指南/书籍医学领域知识库自然语言问答病症搜索推荐诊断搜索推荐构建用药搜索推荐疾病症状用药诊断病例大数据案例支撑分类归档支撑病例数据挖掘n语言是人类思维的工具n语言认知能力是人类作为万物之灵的根本能力机器语言认知是实现类人智能的关键机器能够理解人类语言么？•Ambiguous contextual implicithuman cognitioninfinitePart of materialsare from‘Watson System’by Devendra Chaplot et al.Conceptulization (概念化)Association (联想)Categorization (范畴化)Metaphor (隐喻理解)Frame (场景理解)Induction (归纳)Inference (推理)Xiangyan Sun, et al, Syntactic Parsing of WebQueries.(submitted to EMNLP 2016)Example :cover iphone 6 plus àcoveriphone 6 plus cover àcover576577578579580581582583584585586587588589590591592593594EMNLP 2016Submission ***.Confidential review copy.DO NOT DISQueryParserStanford parsertoys kids NNS NNSnnkids toys NNS NNSnn toys kids NNS NNSnnkids toys NNS NNSnnvanguard school lake wales NN NN NN NNSnnnnnn vanguard school lake wales NN NN NN NNSnnnnnnpretty little liars season 4episode 6RB JJ NNS NN CD NN CD advmod nnnn numnn numpretty little liars season 4episode 6RB JJ NNS NN CD NN CDadvmodnnnn numnnnuminterview questions contract specialistNN NNS NN NNnnnnnncontract specialist interview questionNN NN NN NNnnnnnninterview questions contract specialistNN NNS NN NNnnnnnncontract specialist interview questionNN NN NN NNnnnnnnTable 4:Case study of parsers.5.1Case Study5.3Web Query TreeWe evaluate the 3steps ter each step,we sampl and manually compute We also count the numb each step.The results a•Inferring a depen sentence)pair,w the sentence to th stances shown in ber of (query,sen we obtain depende queries,while the ing criterion.Thi By sacrificing rec high precision.G precision is more•Inferring a uniqu问题:Given words, can we inference the concept these words are talking about?Conceptulization(概念化) Association(联想)Categorization(范畴化)Metaphor(隐喻理解)Frame(场景理解)Induction(归纳) Inference(推理)问题:Find the related entity to complete a minimum conceptExample：baidu, ali->tencentBecause they all belong to concept BAT)Yi Zhang, et al, Entity suggestion with conceptual explanation, (submitted to ICDE2017)Conceptulization (概念化)Association (联想)Categorization (范畴化)Metaphor (隐喻理解)Frame (场景理解)Induction (归纳)Inference (推理)Example ：Sex=man,Marriage status=unmarried àBachelorrich property-value information for entities only.Inferencing DFs of categories from their entities is not easy.In the tra-ditional efforts,psychologists manually construct DFs for a very limited number of popular categories,such as birds ,an-imals ,and cars [Collins and Quillian,1969;Tulving,1972;Smith et al.,1974].However,it is impractical to find the DFs for millions of categories in the real world manually,es-pecially considering that there are millions of tail categories which are unfamiliar to most of us.ANIMAL BIRD OWL [Multicellular, Eukaryotic, Kingdom Animalia][Order Strigiform][feathered, winged, bipedal, warm-blooded,egg-laying, vertebrate]ROBIN FISHSHARKisA isAisA isA isA [red-breast]Figure 1:An example of DFs of categories.The goal of this paper thus is to automatically find defin-ing features for millions of categories in the real world.define a category.For films directedpher Nolan ,we use its basic type film as Feature Mining our solution framework,then elaborate each framework.orkFramework of Defining Feature Mining.a bootstrapping approach to find theDFs of Bpedia.We illustrate its iterative procedure e refer to a category and its DFs as a C-DFstion consists of four major steps.In the first t DFs of categories from DBpedia.In theP (f |c ).The larger this probability,the greater the proportion of category members sharing f and the more predictable the feature set is of category members.Inter-class similarity is afunction of P (c |f ).The larger this probability,the fewer theobjects in contrasting categories that share this feature set and the more predictive the feature set is of the category.Problem Model Theoretically,when f is the DFs of c ,both P (f |c )and P (c |f )equal to 1,and consequently score (c,f )=1.Because as DFs of c ,all entities in c have f and all en tities that have f belong to c .However,this is the idea case.In reality,due to the incompleteness of knowledge base score (c,f )is far less than 1.Because some entities migh miss some features in the knowledge base,which would un derestimate P (f |c ).Some entities might miss some cate gories,which would underestimate P (c |f ).Hence,more re alistically,we expect to find a feature set f which maximize the score:ˆf (c )=arg max f score (c,f )(4Still due to the incompleteness of knowledge bases,some features might be absent in knowledge base.As a result,fo some categories,it is possible that we cannot find a featurese 问题:Categorize an entity by its properties Bo Xu, et al, Learning Defining Features for Categories.(IJCAI 2016)Conceptulization(概念化) Association(联想)Categorization(范畴化)Metaphor(隐喻理解)Frame(场景理解)Induction(归纳)Inference(推理)Hongsong Li, et al, Data-Driven Metaphor Recognition and Explanation, (ACL2013)Example:Juliet is the sun.à<Julit,sun>Your words cut deep. à<words, knife>问题:Recognizing metaphors and identifyingsource-target mappingswhich enables inference for metaphor understand-ing,as we will show next.3.2Acquiring Metaphors mWe acquire an initial set of metaphors m from sim-iles.A simile is afigure of speech that explicitly compares two different things using words such as “like”and“as”.For example,the sentence Life islike a journey is a simile.Without the word“like,”it becomes a metaphor:Life is a journey.This property makes simile an attractivefirst target for metaphor extraction from a large corpus.We usethe following syntactic pattern for extraction:h target i BE/VB like[a]h source i(1)where BE denotes is/are/has been/have been,etc.,VB denotes verb other than BE,and h target i andh source i denote noun phrases or verb phrases.Note that not every extracted pair is a metaphor. Poetry is like an art matches the pattern,but it is nota metaphor because poetry is really an art.We willuse to clean such pairs.Furthermore,due to theLifenot mcounterneralprobtheoall mby dceptintuisimipairsalsoS anverysuchTo1,00these27a in a type2or3metaphor,since a metaphor is an unusual use of x(the target)within a given context. Therefore P(C|x,y)=P(C|y),where P(C|y)is available from Eq.(9).Similarly,we haveP(h|x,C)=P(x,h)P(C|h)P(x,C)(12)where P(x,h)is obtained from H and P(C|h)is from the context preference distribution.To explain the metaphor,or uncover the missing concept,y⇤=arg maxy^(y,x)2 mP(y|x,C)=arg maxy^(y,x)2 mP(y,x)P(C|y) As a concrete example,consider sentence My carConceptulization(概念化) Association(联想)Categorization(范畴化)Metaphor(隐喻理解)Frame(场景理解)Induction(归纳) Inference(推理)问题:Find verb patterns to understand verb-centric frame semanticsExample:Q:hey robot,can you clean in the living room now?à{action:’clean’,location:‘living room’}Wanyun Cui, et al, Verb Pattern: A Probabilistic Semantic Representation on Verbs, (AAAI 2016)Conceptulization (概念化)Association (联想)Categorization (范畴化)Metaphor (隐喻理解)Frame (场景理解)Induction (归纳)Inference (推理)Xiangyan Sun, Yanghua Xiao*, Haixun Wang, Wei Wang, On Conceptual Labeling of a Bag of Words, (IJCAI 2015)Examplebride, groom, dress, celebration → wedding问题:Given a bag of words, can we inference what the article is talking about?Conceptulization (概念化)Association (联想)Categorization (范畴化)Metaphor (隐喻理解)Frame (场景理解)Induction (归纳)Inference (推理)Liang Jiaqing, et al, On the Transitivity Inference in a Data Driven Conceptual Taxonomy, (submitted to AAAI 2017)问题：Can we infer missing facts from existing facts in knowledge base?Example:Can we infer that Steve Jobs is a Billionaire from the fact that Bill gates is a scientist?语言认知搜索更智慧推荐更智能画像更精准交互更自然搜索直接通向答案更准确地理解搜索意图更智能地推荐相关搜索更优化的结果排序•语法解析•搜索分类•关键词扩展•共现推荐•概念推荐•社会化推荐•语义匹配•排序融合•反馈优化结果排序搜索推荐意图理解领域知识库Demo Address ：Code Search@•数据稀疏•隐私保护不全•标签含有噪音•粒度不合适不准基于知识图谱的精准推荐基于知识图谱的标签泛化基于知识图谱的标签扩展基于社交图谱的标签扩展axyb cChrist, 0.33pop, 0.33food, 0.33freedom, 0.25love, 0.25Christ, 0.25fashion, 0.25photography, 0.5fashion, 0.5f xa =0.5f xb =0.2a xybcafter 1 roundmusic, 0.5food, 0.5food, 0.665music, 0.5Christ, 0.24fashion, 0.175pop, 0.165photography, 0.1freedom, 0.075love, 0.075fashion, 0.275photography, 0.2freedom, 0.075love, 0.075Christ, 0.075f xx =1f yb =0.2f xc =0.3f yc =0.3n精准医疗，首先是个性化医疗；个性化医疗的前提是对病人的精准画像n 用户画像技术是将疾病研究拓展到病人的生活维度的关键技术An Integrated Tag Recommendation Algorithm Towards Weibo User Profiling, (DASFAA 2015),Tag Propagation Based Recommendation across Diverse Social Media, (WWW 2014)语义失配•语言多样性、模糊性，使得语义相似的病例、治疗方案有着千差万别的描述冷启动•很多新病例，不存在或存在很少类似诊疗方案，无从推荐推荐的“艺术性”•很多医生对于质量方案的选择出于直觉，或者难以明言的(隐式的)因素.利用知识图谱扩展、规范化实体描述语义相似性，寻找更多的相似病例、方案-+······················································tweet featurespooling convolutionwords non-linear combinationoutputaverageminmaxFigure 2:The context-based neural network moddos Santos and Gatti 2014;Kalchbrenner,Grefenstette,and Blunsom 2014).The baseline model takes word embedding th t i-+········································································tweet features contextualize poolingconvolutionwordsnon-linear combinationoutputmax minaverageminmax Figure 2:The context-based neural network model for Twitter sentimdos Santos and Gatti 2014;Kalchbrenner,Grefenstette,and Blunsom 2014).The baseline model takes word embedding features from the tweet context itself,and performs feature the number of tweets in t i that shouldbe collec Topic-based context利用深度学习捕捉隐式特征Semantic-based Recommendation Across Heterogeneous Domains, (ICDM 2015)A Graph-based Recommendation across Heterogeneous Domains, (CIKM 2015),三个关键技术•问题理解•答案检索•答案排序两个主要挑战•语义表达的多样性•知识库的海量规模Wanyun Cui, Yanghua Xiao*, et al. ...KBQA: An Online Template Based Question Answering System over Freebase, (IJCAI 2016) CCF Rank A ConferenceDemo Address ：KBQA@•结果•从互联网学习了近三千万种问题语义模板。
汉语表主观极量义的成分复现格式“N的N”研究
编 号：硕士学位论文题 目： 汉语表主观极量义的成分复现格式“N 的N”研究培 养 单 位： 文学院 专 业 名 称： 汉语国际教育 指 导 教 师： 夏军 研 究 生： 卢占鳌完 成 时 间： 2018年3月20日沈阳师范大学研究生处制类别全日制研究生 教育硕士 同等学力本文得到了教育部人文社会科学重点研究基地重大项目(13JJD740001；15JJD740002)的支持，谨致谢忱。
汉语表主观极量义的成分复现格式“N的N”研究摘 要本文根据构式语法，从多方面对“N的N”格式进行了归纳和分析，并通过对比分析考察“N的N”格式的缘起，并针对其在对外汉语教学中的应用进行了分析。
全文主要分成六个部分：第一部分是本文的引言部分。
主要叙述了本文的选题缘起以及对该格式进行分析研究的意义，并对相关的已有研究进行了归纳与总结，最后简述了本文的研究方法与思路，对本文的研究对象进行了界定，列出了本文语料的主要来源和对语料的删减情况。
第二部分是对“N的N”格式的结构探讨，主要包括了对“N的N”格式中名词的准入条件的归纳，并对“N的N”结构做了进一步的分类。
第一阶段将“N 的N”分为程度类和非程度类，第二阶段将将程度类分为方位名词类和抽象名词类，第三阶段将抽象名词类分为“N”能被“很”修饰类和“N”能被数量短语修饰类。
最后，对“N的N”格式的句法功能和句类分布做了归纳和总结。
“N的N”结构做主要作主语和宾语，陈述句和感叹句分布较广，问句和祈使句分布较少。
第三部分是对“N的N”结构在语义和语用方面的分析。
首先确定了程度类“N的N”格式可以作为构式看待，构式义为主观极量义。
格式包含比较义、主观义和极量义。
该格式主要出现在口语语境中，表达说话人强烈的主观情绪与观点。
第四部分是对两个相似结构“N中的N”格式和“N的N”格式的对比分析。
首先对“N中的N”的已有研究进行了归纳与总结。
之后，根据语料考察了三种情况：名词只能进入“N中的N”格式的情况，名词只能进入“N的N”格式的情况，以及名词可以同时进入两种格式的情况。
英语词法知识点总结
英语词法知识点总结When learning English as a second language, understanding lexical knowledge is essential for vocabulary acquisition and language fluency. Lexical knowledge refers to the understanding and use of words, phrases, and collocations in context. This includes understanding meanings, grammatical patterns, and usage of words. In this article, we will explore various lexical knowledge points in English, including word formation, word classes, collocations, idioms, and phrasal verbs.Word FormationWord formation in English refers to the ways in which new words are created, either through affixation, compounding, blending, conversion, or borrowing from other languages. Understanding word formation helps learners to expand their vocabulary and comprehend the meaning of new words.Affixation is a common method of word formation, which involves adding prefixes, suffixes, or infixes to a base word to change its meaning or form. For example, the prefix "un-" added to the word "happy" creates the word "unhappy," changing the meaning to the opposite of happy. Suffixes like "-able" and "-less" are also used to form adjectives, such as "comfortable" and "careless."Compounding involves combining two or more words to create a new word with a distinct meaning. For instance, "rain" + "bow" = "rainbow." Compounds can be formed from different word classes, such as noun-noun compounds (e.g., "chocolate cake"), verb-noun compounds (e.g., "homecoming"), and adjective-noun compounds (e.g., "blackboard").Blending is the process of combining parts of two words to create a new word. For example, "brunch" is a blend of "breakfast" and "lunch," while "smog" combines "smoke" and "fog."Conversion, also known as zero derivation, occurs when a word changes its word class without adding any affixes. For example, the noun "hammer" becomes a verb in the sentence "He hammered the nail."Borrowing involves adopting words from other languages into English. For example, "piano" from Italian, "ballet" from French, and "safari" from Swahili.Word ClassesIn English, words are categorized into different word classes, such as nouns, verbs, adjectives, adverbs, pronouns, prepositions, conjunctions, and interjections. Understanding word classes is essential for sentence structure, grammar, and comprehension.Nouns are words that represent people, places, things, or ideas. They can be concrete (e.g., "table," "dog") or abstract (e.g., "love," "truth").Verbs express actions (e.g., "run," "eat") or states (e.g., "be," "have") and are used to form sentences and express tense, aspect, and mood.Adjectives modify or describe nouns and pronouns (e.g., "beautiful," "tall").Adverbs modify verbs, adjectives, or other adverbs and provide information about time, manner, place, frequency, or degree (e.g., "slowly," "very").Pronouns replace nouns or noun phrases (e.g., "he," "she," "it").Prepositions show the relationship between a noun or pronoun and other words in a sentence (e.g., "in," "on," "at").Conjunctions connect words, phrases, or clauses (e.g., "and," "but," "or").Interjections express emotions or feelings (e.g., "wow," "ouch," "oh").CollocationsCollocations are words that often occur together and form natural combinations in language. Understanding collocations helps learners use words in context and produce natural-sounding language.There are different types of collocations, such as adjective-noun collocations (e.g., "heavy rain"), verb-noun collocations (e.g., "make a decision"), adverb-adjective collocations (e.g., "extremely happy"), and noun-preposition collocations (e.g., "aware of").IdiomsIdioms are expressions that have a figurative meaning different from the literal meaning of the individual words. Understanding idioms is essential for comprehension and communicating fluently in English.Common idioms include "break the ice" (to initiate social interaction), "a piece of cake" (something that is easy), and "hit the hay" (to go to bed).Phrasal VerbsPhrasal verbs are combinations of a verb and one or more particles (adverbs, prepositions, or both) that have idiomatic meanings. Understanding phrasal verbs is important for language fluency and natural communication.Examples of phrasal verbs include "give up" (to stop doing something), "get along" (to have a good relationship), and "look after" (to take care of).In conclusion, lexical knowledge plays a crucial role in English language learning and usage. Understanding word formation, word classes, collocations, idioms, and phrasal verbs is essential for vocabulary acquisition, comprehension, and fluency. By mastering these lexicalknowledge points, learners can improve their language skills and communicate effectively in English.。
提取文本英语单词
提取文本英语单词Extracting English words from text is a fundamental taskin various fields such as natural language processing,machine learning, and data analysis. This process involves identifying and separating individual words from a largerbody of text to facilitate further analysis or manipulation.One common approach to extracting English words isthrough tokenization. Tokenization is the process ofsplitting a string of text into individual elements, known as tokens, which are typically words or phrases. This can bedone using various algorithms and tools that recognize word boundaries and punctuation marks.For instance, in programming, one might use a regular expression (regex) to split a text string into words. Asimple regex pattern like "\W+" can be used to split a string at every non-word character, effectively extracting the words.Another method is to use natural language processing libraries available in many programming languages. For example, the Natural Language Toolkit (NLTK) in Pythonprovides a comprehensive suite of tools for working with human language data. It includes functions for tokenization, along with many other features for text processing.Once the words are extracted, they can be used for a multitude of purposes. For example, they can be counted tocreate word frequency distributions, which can be insightful for understanding the content and context of the text. Additionally, extracted words can be used to perform sentiment analysis, where the emotional tone behind words is examined to determine the writer's attitude or feelings.Furthermore, in the field of machine learning, extracted words can be used as features for training models to perform tasks such as text classification or topic modeling. These models can then be used to categorize new texts or to discover latent topics within a corpus of documents.In conclusion, the extraction of English words from text is a crucial step in many text-based applications. It allows for the transformation of continuous text into discrete, analyzable units, opening the door to a wide range of analytical techniques and applications.。
Corpus stylistics
AS IF IT WERE A THE OPPOSITE SIDE OF THE
A QUARTER OF AN HOUR AT THE BOTTOM OF THE
ON THE PART OF THE WHAT DO YOU THINK OF
systematic distinctions meant to account for the whole of a language (local grammars) • lexical items have textual functions
Corpus stylistics?
Corpus linguistics: repetition, norms, the typical
Cf. Conrad & Biber (2005) discourse functions of lexical bundles, Mahlberg (2005) textual function of general noun patterns
Examples from literary texts
(THE) UNITED METROPOLITAN IMPROVED HOT MUFFIN AND CRUMPET BAKING AND PUNCTUAL DEVLIVERY COMPANY (13 words, NN)
8 (+) clusters in 19C: 30 8-word clusters (min 5)
A QUARTER OF AN HOUR AT THE END OF THE
THE OTHER SIDE OF THE IN THE MIDDLE OF THE IN THE COURSE OF THE
ROBUST INFORMATION EXTRACTION FROM SPOKEN LANGUAGE DATA
1. INTRODUCTION
Extracting linguistic structure such as proper names, noun phrases, and verb phrases is an important rst step in many systems aimed at automatic language understanding. While signi cant progress has been made on this problem, most of the work has focused on \clean" textual data such as newswire texts, in which cues such as capitalization and punctuation are important for obtaining high accuracy results. However, there are many data sources where these cues are not reliable, such as in spoken language data or single-case text. Spoken language sources in particular pose additional problems because of dis uencies and speech recognition errors. Previous approaches that have addressed spoken language have consisted largely of applying an existing text-based system to speech data, ignoring the fact that information lost due to recognition errors when moving from text to speech can be regained in part via word con dence prediction. In 3], we introduced a new probabilistic model for the problem of information extraction (IE) for spoken language data and demonstrated state-of-the-art results, even for high error rate tasks. We will review the model in Section 2. One signi cant contribution of our system is its robustness in processing errorful speech transcriptions. In this paper we explore the robustness issue further by addressing an important problem in developing IE systems for noisy data: mismatch between the training data and the test data. For speech data, in particular, the training data consists of \clean" human transcriptions of spoken language, while the test data
Noun Compounds
The three written forms
• The open or spaced form consisting of newer combinations of usually longer words, such as distance learning, player piano, lawn tennis, etc.
Further Information
• The three written forms & Usage • Sound patterns (stress) • Plural forms of compound nouns
The three written forms
• The "solid" or "closed" forms, in which
singular
a school teacher one assistant headmaster the sergeant major a mother-in-law an assistant secretary of state my toothbrush a doctor of philosophy a passerby, a passer-by
Noun Compounds
Noun Compounds
• • • • summary Sub classification string compound compound nouns formed from phrasal verbs • Further information
ቤተ መጻሕፍቲ ባይዱ
Summary
usage
• Usage in the US and in the UK differs and often depends on the individual choice of the writer; therefore, open, hyphenated, and closed forms may be encountered for the same compound noun. It is partly a matter of style. There are no definite rules.