BibSonomyburst/user/diego_ma/questionsBibSonomy RSS feed for /user/diego_ma/questions2012-02-16T22:16:05+01:00http://www.bibsonomy.org/bibtex/207a43efda03f9e9c24cf9fe818a1f4e6/diego_madiego_ma2011-02-07T09:47:40+01:00questions <span class="authorEditorList"><a href="/author/Hallett">Catalina Hallett</a>, <a href="/author/Power">Richard Power</a>, and <a href="/author/Scott">Donia Scott</a> </span><em>Computational Linguistics</em> (<em>2007</em>)Mon Feb 07 09:47:40 CET 2011Computational Linguistics1Composing Questions through Conceptual Authoring332007questions http://www.bibsonomy.org/bibtex/2c1668fd0f8578440a8ec619d9e02b46e/diego_madiego_ma2009-07-22T09:58:28+02:00questions question_answering biomedical <span class="authorEditorList"><a href="/author/Terol">Rafael M. Terol</a>, <a href="/author/Martínez-Barco">Patricio Martínez-Barco</a>, and <a href="/author/Palomar">Manuel Palomar</a> </span><em>MICAI 2006: Advances in Artificial Intelligence, </em><em>volume 4293 of Lecture Notes in Computer Science, </em><em>page 996-1006. </em><em>Springer, </em>(<em>2006</em>)Wed Jul 22 09:58:28 CEST 2009MICAI 2006: Advances in Artificial Intelligence996-1006Lecture Notes in Computer ScienceApplying NLP Techniques and Biomedical Resources to Medical Questions in QA Performance.42932006questions question_answering biomedical Nowadays, there is an increasing interest in research on QA over restricted domains. Concretely, in this paper we will show the process of question analysis in a medical QA system. This system is able to obtain answers to different natural language questions according to a question taxonomy. In this system we combine the use of NLP techniques and biomedical resources. The main NLP technique is the use of logic forms and the pattern matching technique in this question analysis performance.dblphttp://www.bibsonomy.org/bibtex/23cb7b8d122f144ccf7957fac478a3ed2/diego_madiego_ma2008-03-04T07:47:30+01:00questions machine_learning <span class="authorEditorList"><a href="/author/Pizzato">Luiz A.S. Pizzato</a> </span><em>Proc. ALTW 2004, </em><em>page 25-31. </em><em>Sydney, Australia, </em><em>Macquarie University, </em><em>ASSTA, </em>(<em>2004</em>)Tue Mar 04 07:47:30 CET 2008Sydney, AustraliaProc. ALTW 200425-31Using a Trie-based Structure for Question Analysis2004questions machine_learning This paper presents an approach for question analysis that defines the question subject and its required answer type by building a trie-based structure from a set of question patterns. The question analysis consists of comparing the question tokens with the path of nodes in the trie. A look-ahead process solve the mismatches of unknown words by assigning a entity-type or semantically linking them with other question words. The developed approach is evaluated using different datasets showing that its performance is comparable with state-of-the-art systems.http://www.bibsonomy.org/bibtex/21753d3843e3842b957bb0a0660ab4513/diego_madiego_ma2008-02-19T06:58:42+01:00biomedical questions clinical databases <span class="authorEditorList"><a href="/author/Alper">Brian S. Alper</a>, <a href="/author/Stevermer">James J. Stevermer</a>, <a href="/author/White">David S. White</a>, and <a href="/author/Ewigman">Bernard G. Ewigman</a> </span><em>Journal of Family Practice</em> <em>50(11):960-965</em> (<em>2001</em>)Tue Feb 19 06:58:42 CET 2008Journal of Family Practice11960-965Answering Family Physicians' Clinical Questions Using Electronic Medical Databases502001biomedical questions clinical databases KEY POINTS FOR CLINICIANS * Five current electronic medical databases can answer 50% or more of family physicians' clinical questions. * Combinations of currently available databases can answer 75% or more of these questions. * Getting answers to clinical questions can often take from 2 to 6 minutes for each query. * OBJECTIVE We studied the ability of electronic medical databases to provide adequate answers to the clinical questions of family physicians. * STUDY DESIGN Two family physicians attempted to answer 20 questions with each of the databases evaluated. The adequacy of the answers was determined by the 2 physician searchers, and an arbitration panel of 3 family physicians was used if there was disagreement. * DATA SOURCE We identified 38 databases through nominations from national groups of family physicians, medical informaticians, and medical librarians; 14 of these databases met predetermined eligibility criteria. * OUTCOME MEASURED The primary outcome was the proportion of questions adequately answered by each database and by combinations of databases. We also measured mean and median times to obtain adequate answers for individual databases. * RESULTS The agreement between family physician searchers regarding the adequacy of answers was excellent ([kappa]=0.94). Five individual databases (STAT!Ref, MDConsult, DynaMed, MAXX, and MDChoice.com) answered at least half of the clinical questions. Some combinations of databases answered 75% or more. The average time to obtain an adequate answer ranged from 2.4 to 6.5 minutes. * CONCLUSIONS Several current electronic medical databases could answer most of a group of 20 clinical questions derived from family physicians during office practice. However, point-of-care searching is not yet fast enough to address most clinical questions identified during routine clinical practice.http://www.bibsonomy.org/bibtex/24bcc088fa8af85007e475d2c0c90348d/diego_madiego_ma2007-12-14T02:48:40+01:00questions question_answering <span class="authorEditorList"><a href="/author/Zhang">Dell Zhang</a>, and <a href="/author/Lee">See Sun Lee</a> </span><em>Proc. SIGIR 03, </em><em>ACM, </em>(<em>2003</em>)Fri Dec 14 02:48:40 CET 2007Proc. SIGIR 03Question Classification using Support Vector Machines2003questions question_answering Question classification is very important for question answering. This paper presents our research work on automatic question classification through machine learning approaches. We have experimented with five machine learning algorithms: Nearest Neighbors (NN), Na?ve Bayes (NB), Decision Tree (DT), Sparse Network of Winnows (SNoW), and Support Vector Machines (SVM) using two kinds of features: bag-of-words and bag-ofngrams. The experiment results show that with only surface text features the SVM outperforms the other four methods for this task. Further, we propose to use a special kernel function called the tree kernel to enable the SVM to take advantage of the syntactic structures of questions. We describe how the tree kernel can be computed efficiently by dynamic programming. The performance of our approach is promising, when tested on the questions from the TREC QA track.http://www.bibsonomy.org/bibtex/2a2db363e44bd0ae79fe4da8efb3b4a5c/diego_madiego_ma2007-12-14T02:46:30+01:00semantics questions <span class="authorEditorList"><a href="/author/Shan">Chung-Chieh Shan</a>, and <a href="/author/ten Cate">Balder ten Cate</a> </span><em>Proc. 7th ESSLLI Student Session, </em>(<em>2002</em>)Fri Dec 14 02:46:30 CET 2007Proc. 7th ESSLLI Student SessionThe Partition Semantics of Questions, Syntactically2002semantics questions Groenendijk and Stokhof (1984, 1996; Groenendijk 1999) provide a logically attractive theory of the semantics of natural language questions, commonly referred to as the partition theory. Two central notions in this theory are entailment between questions and answerhood. For example, the question ``Who is going to the party?'' entails the question ``Is John going to the party?'', and ``John is going to the party'' counts as an answer to both. Groenendijk and Stokhof define these two notions in terms of partitions of a set of possible worlds. We provide a syntactic characterization of entailment between questions and answerhood . We show that answers are, in some sense, exactly those formulas that are built up from instances of the question. This result lets us compare the partition theory with other approaches to interrogation -- both linguistic analyses, such as Hamblin's and Karttunen's semantics, and computational systems, such as Prolog. Our comparison separates a notion of answerhood into three aspects: equivalence (when two questions or answers are interchangeable), atomic answers (what instances of a question count as answers), and compound answers (how answers compose).http://www.bibsonomy.org/bibtex/2e620a713b0d1f6cb56a4333e8d8c6cf4/diego_madiego_ma2007-12-14T02:45:02+01:00questions machine_learning <span class="authorEditorList"><a href="/author/Pizzato">Luiz A.S. Pizzato</a> </span><em>Proc. ALTW 2004, </em><em>page 25-31. </em><em>Sydney, Australia, </em><em>Macquarie University, </em><em>ASSTA, </em>(<em>2004</em>)Fri Dec 14 02:45:02 CET 2007Sydney, AustraliaProc. ALTW 200425-31Using a Trie-based Structure for Question Analysis2004questions machine_learning This paper presents an approach for question analysis that defines the question subject and its required answer type by building a trie-based structure from a set of question patterns. The question analysis consists of comparing the question tokens with the path of nodes in the trie. A look-ahead process solve the mismatches of unknown words by assigning a entity-type or semantically linking them with other question words. The developed approach is evaluated using different datasets showing that its performance is comparable with state-of-the-art systems.http://www.bibsonomy.org/bibtex/24c026bf9c5b2977307ca3b642436af13/diego_madiego_ma2007-12-14T02:42:24+01:00question_answering questions <span class="authorEditorList"><a href="/author/Li">Xin Li</a>, and <a href="/author/Roth">Dan Roth</a> </span><em>Proc. COLING 02</em> (<em>2002</em>)Fri Dec 14 02:42:24 CET 2007Proc. COLING 02Learning Question Classifiers2002question_answering questions In order to respond correctly to a free form factual question given a large collection of texts, one needs to understand the question to a level that allows determining some of the constraints the question imposes on a possible answer. These constraints may include a semantic classification of the sought after answer and may even suggest using different strategies when looking for and verifying a candidate answer. This paper presents a machine learning approach to question classification. We learn a hierarchical classifier that is guided by a layered semantic hierarchy of answer types, and eventually classifies questions into fine-grained classes. We show accurate results on a large collection of free-form questions used in TREC 10.http://www.bibsonomy.org/bibtex/27bc1e7d4b370103c868c43fcc90face7/diego_madiego_ma2007-12-14T02:37:17+01:00semantics questions prolog <span class="authorEditorList"><a href="/author/ten Cate">Balder ten Cate</a>, and <a href="/author/chieh Shan">Chung chieh Shan</a> </span><em>Proc. TABLEAUX 2002, </em><em>page 251-265. </em>(<em>2002</em>)<em>Also in Proc. NLULP 2002 . </em>Fri Dec 14 02:37:17 CET 2007Proc. TABLEAUX 2002Also in Proc. NLULP 2002251-265Lecture Notes in Artificial Intelligence 2381Question Answering: From Partitions to Prolog2002semantics questions prolog We implement Groenendijk and Stokhof's partition semantics of questions in a simple question answering algorithm. The algorithm is sound, complete, and based on tableau theorem proving. The algorithm relies on a syntactic characterization of answerhood: Any answer to a question is equivalent to some formula built up only from instances of the question. We prove this characterization by translating the logic of interrogation to classical predicate logic and applying Craig's interpolation theorem.http://www.bibsonomy.org/bibtex/23e69e85e9664f417bc680eb6f6c56c6c/diego_madiego_ma2007-12-14T02:36:22+01:00semantics questions <span class="authorEditorList"><a href="/author/Bos">Johan Bos</a>, and <a href="/author/Gabsdil">Malte Gabsdil</a> </span><em>Proc. Goetalog 2000, </em><em>page 43-50. </em>(<em>2000</em>)Fri Dec 14 02:36:22 CET 2007Proc. Goetalog 200043-50Papers in Computational Linguistics 00-5First-Order Inference and the Interpretation of Questions and Answers2000semantics questions Building on work by Groenendijk and Stokhof, we develop a theory of question and answer interpretation for first-order formalisms. The proposed framework is less fine-grained than its higher-order ancestor, but instead offers attractive implementational properties as it deals with the combinatorial explosion problem underlying Groenendijk and Stokhof?s original theory. To incorporate the treatment of questions and answers in a larger setting, we use an extension of Discourse Representation Theory to cover typical contextual phenomena such as anaphora and presupposition. The actual interpretation of the dialogue representation is done via a translation to first-order logic. A prototype implementation, using state-of-the-art theorem proving and model building facilities, supports the idea that this first-order approximation of the interpretation of questions and answers is indeed a useful one.