... semi -supervised learning for the question- answering (QA)task for ranking candidate sentences. Us-ing textual entailment analysis, we obtainentailment scores between a natural lan-guage question ... tex-tual entailment analysis. Question- Type Classifier (QC): QC is the taskof identifying the type of a given question amonga predefined set of question types. The type ofa question is used as a clue ... methods.line represents a converted question, in order toextract the question- type feature, we use a match-ing NER-type between the headline and candidatesentence to set question- type NER match feature.We...
... rating”want in” answer ”in the” in question CA avg asker rating”adenocarcinoma” in question CA History “anybody” in question ”was” in question ”who is” in question Q content typo density”live” ... exist to automat-ically answer open domain, complex, and subjectivequestions. Hence, this paper contributes to both theunderstanding of complexquestion answering, andexplores evaluation issues ... Figure 1.• Question Features: Traditional question answer-ing features such as the wh-type of the question (e.g., “what” or “where”), and whether the ques-tion is similar to other questions...
... time. The complexity and length ofa query on the other hand has very little impact onspeed.The type of question influences the query buildingprocess in a fundamental manner. For the question “When ... for differ-ent type of questions, increases or decreases con-stituents’ weights in either the Headers or the Textfield.Additionally, during question analysis, certain question constituents are ... auser’s search experience.1 IntroductionThis paper describes the online demo ofthe QuALiM1 Question Answering system(http://demos.inf.ed.ac.uk:8080/qualim/). Wewill refrain from describing...
... list of 20 questions for each language. Finally, we produced the Gold Standard, by labeling the corpus with the correct answers corresponding to the questions. 1.1 Questions No TYPE QUESTION ... containing NEs of the question Expected Answer Type (EAT). Table 2 presents the results ob-tained for English and Table 3 for Spanish. We indicate the id of the question (Q), the question type (T) ... annotate it or wrongfully an-notated words as being NEs. Not of less impor-tance is the question duality aspect in question 17. Bush is commented in more than 600 sen-tences; therefore, when polarity...
... ex-traction for clinical question answering: Preliminaryresults. In AAAI 2005 Workshop on QA in RestrictedDomains.D. Demner-Fushman and J. Lin. 2006, in press. An-swering clinical questions with knowledge-basedand ... and Teach EBM.Churchill Livingstone, second edition.E. Voorhees. 2005. Using question series to eval-uate questionanswering system effectiveness. InHLT/EMNLP 2005.Y. Zhao and G. Karypis. 2002. ... hybrid approachto questionanswering in the clinicaldomain that combines techniques fromsummarization and information retrieval.We tackle a frequently-occurring class ofquestions that takes...
... for generat-ing encyclopedias is of operational quality.5.3 Evaluating Question Answering We used as test inputs 40 questions, which are relatedto technical terms collected from the Class ... domains (and consequently word senses).In addition, we proposed a questionanswering sys-tem, which answers interrogative questions associatedwith what, by using a Web-based encyclopedia as ... ofoperational quality and quantity.We also used test questions from the Class II exam-ination, and evaluated the Web-based encyclopedia interms of question answering. We found that our Web-based encyclopedia...
... Scha, W.J, Schoenmakers and E.P.C. van Utteren. The QuestionAnswering System PHLIQA1. In L. Bolc (editor), Natural Language Question Answering Systems. Macmillan, 1980. [8] Scha, Remko J.H. ... mapping is a component of such question- answering systems as TEAM [4], PHLIQA1 [7] and IRUS [1]. As an example of difficulties which may be encountered, consider the question "Was the patient's ... particular boolean field which encodes for each patient whether or not this complex property is true. Any variation on this question which a compositional semantics might also handle, such as "Was...
... knowledge and external resources forTREC-style question answering. In particular, wedescribe a high performance question answering system called QUALIFIER (QUestion Answer-ing by Lexical Fabric and ... and its Applicability to Question Answering. In Proceedings of the ACL2001 Conference, July 2001.M. A. Pasca and S. M. Harabagiu. 2001. High per-formance question/ answering. In Proceedings ... 2002. Probabilistic question answering from the web. In The Eleventh Interna-tional World Wide Web Conference,2002.E.M.Voorhees. 2002. Overview of the TREC 2001 Question Answering Track. In...
... post. Only identifying question and answering sentences from the thread is not enoughfor automatic question answering. For this example,in order to get the complete question, we would needto ... online forums.1 IntroductionAutomatic QuestionAnswering (QA) systems relyheavily on good sources of data that contain ques-tions and answers. Questionanswering forums, suchas technical support ... Example of a QuestionAnswering Thread inUbuntu Support Forumexists as a result of another sentence. For example, question context sentences exist because of the ques-tion itself; an answering...
... factoid questions in TREC 2004 has230 questions. Lexical chains containing the re-stricted set of relations that propagate verb argu-ments were found for 33 questions, linking verbsin those questions ... Lexicalchains for question answering. In Proceedings ofCOLING 2002, pages 674–680.Dan I. Moldovan and Vasile Rus. 2001. Logic FormTransformation of WordNet and its Applicability to Question Answering. ... system performance by 2.4%.1 IntroductionIn QuestionAnswering the correct answer can beformulated with different but related words thanthe question. Connecting the words in the ques-tion...
... Paragraphs Question DocumentsAnswersAnswers−M1Answers−M2Answers−M3 QUESTION ANSWERING SYSTEMFigure 1: Integrating Textual Entailment in Q/A.lows. Section 2 describes the three methods ofusing textual entailment in open-domain question answering ... bythe original question, the answer passage associ-ated with the top-ranked entailed question was re-turned as the answer. When none of the top 50AGQs were entailed by the question, question- answer ... 1TEXTUALENTAILMENTMethod 2List of QuestionsGenerationAUTO−QUABRanked List of ParagraphsTEXTUALENTAILMENTMethod 3Entailed QuestionsEntailed ParagraphsList of Entailed Paragraphs Question DocumentsAnswersAnswers−M1Answers−M2Answers−M3QUESTION...
... Dialogue Question AnswerDecomposition Question InformationExtractionRepresentationOff−line Question Answering Database (QUAB) Question AnswerFigure 1: FERRET - A Predictive Interactive Question- Answering Architecture.model, ... theoriginal question and the predicted questions.The selection of the questions from the QUABsthat are proposed for each user question is based ona similarity-metric that ranks the QUAB questions.To ... similarity between the user question and the QUAB question to be definedas , whereSimilarity Metric 4 is based on the question type similarity. Instead of using the question class, determined...
... IntroductionThe conventional QuestionAnswering (QA) archi-tecture is a cascade of the following building blocks: Question Analyzer analyzes a question sentenceand identifies the question types (or answertypes).Document ... Japanyutaka.sasaki@atr.jpAbstractThis paper regards Question Answering (QA) as Question- Biased Term Extraction(QBTE). This new QBTE approach lib-erates QA systems from the heavy bur-den imposed by question types (or answertypes). ... 10,000 pairsof question and answers. You have to manually clas-sify the questions according to your own question- type system. In addition, you have to annotate thetags of the question types...
... efficiently answer questions. We evaluate our strategy on a challenging subset of questions, i.e. “Who is …” questions, against a state of the art web-based Question Answering system. Results ... look-up method’s answer is counted as incorrect. Offline Strategies for Online Question Answering: Answering Questions Before They Are Asked Michael Fleischman, Eduard Hovy, Abdessamad ... concept, and instance. 5 Question Answering Evaluation A large number of questions were collected over the period of a few months from www.askJeeves.com. 100 questions of the form “Who...
... that are trained on question- answer corpora. From this perspective then, the fundamental problem of questionanswering is that of finding spaces where the distance between questions and sentences ... into questions. 4.2 Test Data We used two different data sets for the purpose of testing. The first set consists of the 500 questions used at TREC 2002; the second set consists of 500 questions ... databases were proved to be very useful for question- answering systems. Lin (2002) showed through his federated approach that 47% of TREC-2001 questions could be answered using Web-based...