Answer Extraction

In real world computing environment with using a computer to answer questions has been a human dream since the beginning of the digital era, Question-answering systems are referred to as intelligent systems, that can be used to provide responses for the questions being asked by the user based on certain facts or rules stored in the knowledge base it can generate answers of questions asked in natural , and the first main idea of fuzzy logic was to working on the problem of computer understanding of natural language, so this survey paper provides an overview on what Question-Answering is and its system architecture and the possible relationship and different with fuzzy logic, as well as the previous related research with respect to approaches that were followed. At the end, the survey provides an analytical discussion of the proposed QA models, along or combined with fuzzy logic and their main contributions and limitations.

In this paper we describe ExtrAns, an answer extraction system. Answer extraction (AE) aims at retrieving those exact passages of a document that directly answer a given user question. AE is more ambitious than information retrieval and information extraction in that the retrieval results are phrases, not entire documents, and in that the queries may be arbitrarily specific. It is less ambitious than full-fledged question answering in that the answers are not generated from a knowledge base but looked up in the text of documents. The current version of ExtrAns is able to parse unedited Unix "man pages", and derive the logical form of their sentences. User queries are also translated into logical forms. A theorem prover then retrieves the relevant phrases, which are presented through selective highlighting in their context.

Nowadays there is a big amount of biomedical literature which uses complex nouns and acronyms of biological entities thus complicating the task of retrieval specific information. The Genomics Track works for this goal and this paper describes the approach we used to take part of this track of TREC 2007. As this is the first time we participate in this track, we configurated a new system consisting of the following diferenciated parts: preprocessing, passage generation, document retrieval and passage (with the answer) extraction. We want to call special attention to the textual retrieval system used, which was developed by the University of Alicante. Adapting the resources for the propouse, our system has obtained precision results over the mean and median average of the 66 official runs for the Document, Aspect and Passage2 MAP; and in the case of Passage MAP we get nearly the median and mean value. We want to emphasize we have obtained these results without incorporating specific information about the domain of the track. For the future, we would like to further develop our system in this direction.

Penemuan tentang biologi menyebabkan semakin banyaknya objek yang harus dipelajari dari Biologi, sehingga seorang ilmuwan tidak sanggup lagi mempelajari secara mendalam seluruh objek kajian Biologi. Maka dari itu, cabang-cabang Biologi dikembangkan untuk mempermudah mempelajari masing- masing kajian objek Biologi. Di dalam modul ini akan dijelaskan ruang lingkup biologi (permasalahan pada berbagai objek biologi dan tingkat organisasi kehidupan), melalui penerapan metode ilmiah dan prinsip keselamatan kerja

This paper introduces an approach to mapping relational database contents to ontologies. The current effort is motivated by the need of including into the Semantic Web volumes of web data not satisfied by current search engines. A graphical tool is developed in order to ease the mapping procedure and export enhanced ontologies linked to database entities. Moreover, queries using semantic web query languages can be imposed to the database through its connection to an ontology. Using Protégé, we were able to map ontology instances into relational databases and retrieve results by semantic web query languages. The key idea is that, instead of storing instances along with the ontology terminology, we can keep them stored in a database and maintain a link to the dataset. Thus, on one hand we achieve smaller size ontologies and on the other hand we can exploit database contents harmonizing the semantic web concept with current widespread techniques and popular applications.

Traditionally, the task of answering natural language questions has involved a keyword-based document retrieval step, followed by in-depth processing of candidate answer documents and paragraphs. This post-processing uses semantics to various degrees. In this article, we describe a purely semantic question answering (QA) framework for large document collections. Our high-precision approach transforms the semantic knowledge extracted from natural language texts into a language-agnostic RDF representation and indexes it into a scalable triplestore. In order to facilitate easy access to the information stored in the RDF semantic index, a user's natural language questions are translated into SPARQL queries that return precise answers back to the user. The robustness of this framework is ensured by the natural language reasoning performed on the RDF store, by the query relaxation procedures, and the answer ranking techniques. The improvements in performance over a regular free text search index-based question answering engine prove that QA systems can benefit greatly from the addition and consumption of deep semantic information.

Traditionally, the task of answering natural language questions has involved a keyword-based document retrieval step, followed by in-depth processing of candidate answer documents and paragraphs. This post-processing uses semantics to various degrees. In this article, we describe a purely semantic question answering (QA) framework for large document collections. Our high-precision approach transforms the semantic knowledge extracted from natural language texts into a language-agnostic RDF representation and indexes it into a scalable triplestore. In order to facilitate easy access to the information stored in the RDF semantic index, a user's natural language questions are translated into SPARQL queries that return precise answers back to the user. The robustness of this framework is ensured by the natural language reasoning performed on the RDF store, by the query relaxation procedures, and the answer ranking techniques. The improvements in performance over a regular free text s...

In this article, we present a factoid question-answering system, Sibyl, specifically tailored for question answering (QA) on spoken-word documents. This work explores, for the first time, which techniques can be robustly adapted from the usual QA on written documents to the more difficult spoken document scenario. More specifically, we study new information retrieval (IR) techniques designed or speech, and utilize several levels of linguistic information for the speech-based QA task. These include named-entity detection with phonetic information, syntactic parsing applied to speech transcripts, and the use of coreference resolution. Sibyl is largely based on supervised machine-learning techniques, with special focus on the answer extraction step, and makes little use of handcrafted knowledge. Consequently, it should be easily adaptable to other domains and languages. Sibyl and all its modules are extensively evaluated on the European Parliament Plenary Sessions English corpus, compa...

Este articulo presenta un sistema de Extraccion de Respuestas que opera mediante la transformacion de documentos y preguntas en una representacion semantica denominada Forma Logica Minima (FLM). Un tratamiento especial de la terminologia permite adecuar el sistema a dominios tecnicos.

This paper describes the participation of the Technical University of Catalonia in the CLEF 2009 Question Answering on Speech Transcripts track. We have participated in the English and Spanish scenarios of QAst. For both manual and automatic transcripts we have used a robust factual Question Answering that uses minimal syntactic information. We have also developed a NERC designed to handle automatic transcripts. We perform a detailed analysis of our results and draw conclusions relating QA performance to word error rate and the difference between written and spoken questions.

Dalam kehidupan sehari-hari, Kalian tentu tidak asing lagi dengan bahan makanan, seperti beras, jagung, daging tempe, dan tahu. Kalian juga pasti mengenal berbagai jenis plastik. Berbagai perabotan rumah tangga, seperti ember, gayung, dan gelas yang terbuat dari plastik. Bahan-bahan tersebut terbuat dari senyawa makromolekul atau polimer. Sesuai dengan nama senyawa tersebut, makromolekul berarti molekul besar dengan berat molekul yang besar. Tubuh manusia juga terdiri atas makromolekul- makromolekul yang Kalian kenal dengan karbohidrat, protein, dan lemak. Sekarang, Kalian pikirkan, apakah persamaan dan perbedaan antara bahan makanan dan plastik? Apakah makromolekul dan polimer itu? Bagaimana sifat-sifat makromolekul?, Senyawa apa sajakah yang termasuk makromolekul?, Dengan mempelajari modul ini kalian akan mengetahui jawabannya

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

We present improvements and modifications of the QED open-domain question answering system developed for TREC-2003 to make it cross-lingual for participation in the Cross-Linguistic Evaluation Forum (CLEF) Question Answering Track 2004 for the source languages French and German and the target language English. We use rule-based question translation extended with surface pattern-oriented pre-and post-processing rules for question reformulation to create and English query from its French or German original. Our system uses deep processing for the question and answers, which requires efficient and radical prior search space pruning. For answering factoid questions, we report an accuracy of 16% (German to English) and 20% (French to English), respectively.

This report describes a new open-domain answer retrieval system developed at the University of Edinburgh and gives results for the TREC-12 question answering track. Phrasal answers are identified by increasingly narrowing down the search space from a large text collection to a single phrase. The system uses document retrieval, query-based passage segmentation and ranking, semantic analysis from a wide-coverage parser, and a unification-like matching procedure to extract potential answers. A simple Web-based answer validation stage is also applied. The system is based on the Open Agent Architecture and has a parallel design so that multiple questions can be answered simultaneously on a Beowulf cluster.

Performance on a question answering system is reflected by the number of correct answers that is produce from a list of alternate choices. However, many answers would appear in documents and passages laden with terms from the query. It becomes more difficult and eventually impossible to make a precise answer to a query. The aim of this study is to determine the keyword of the precise answer to a specific query and illustrate the answer in visual expression. In this paper we propose a method to translate text into visual expression, closest to the correct answer. To achieve a significant illustration capability it is necessary to develop methods of dealing with knowledge base. Firstly, we must consider the real world and the visual expression of key information that can be extracted from the words. The proposed system will analyse and match them using visual semantic simplification description based on logical representation.

The idea of the computer system capable of simulating understanding with respect to reading a story or passage and answering questions pertaining to it, has received increased attention within the NLP community as a means to develop and evaluate robust question answering methods. This research is concerned with the problem of generating an automated answer in the context of sophisticated knowledge representation, reasoning, and logical inferential processing. The research focused on Wh types of question to a restricted domain. External knowledge sources i.e. world knowledge (WK) and hypernyms matching procedure (HMP) were introduced. World knowledge led to the refinement of the ability of the system to extract the relevant answers. It provided a solution to the outstanding problem related to the ambiguity enclosed by anaphora and synonym words. Meanwhile, hypernyms refer to a more general or broad keywords, which was able to widen the results of the search. Thus, hypernyms matching procedure was to give a more coherent meaning of words, and eased in the process of extracting the answer based on a given question. This research found that the combined external knowledge sources on word dependencies improved the accuracy of the question answering system with respect to human performance measures.

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

In this paper we argue that questionanswering (QA) over technical domains is distinctly different from TREC-based QA or Web-based QA and it cannot benefit from data-intensive approaches. Technical questions arise in situations where concrete problems require specific answers and explanations. Finding a justification of the answer in the context of the document is essential if we have to solve a real-world problem. We show that NLP techniques can be used successfully in technical domains for high-precision access to information stored in documents. We present Extr-Ans, an answer extraction system over technical domains, its architecture, its use of logical forms for answer extractions and how terminology extraction becomes an important part of the system.

Nous exposons ici l'implémentation d'un système d'extraction de réponses, ExtrAns, qui utilise la sortie d'un analyseur et d'une grammaire basés sur les dépendances. Afin d'augmenter la vitesse de calcul, l'analyseur et la grammaire utilisés sacrifient le fonctionnalisme (dans le cadre des grammaires de dépendance) au profit de la projectivité. Nous avons découvert que les structures de dépendance résultantes, bien qu'elles soient difficiles à traiter, peuvent être utilisées par ExtrAns pour trouver les dépendances syntaxiques et sémantiques nécessaires dans plusieurs des étapes du traitement linguistique. En particulier, nous mettons l'accent sur la génération de formes logiques minimales. We report on the implementation of an answer extraction system, ExtrAns, that uses the output of a dependency-based parser and grammar. In order to increase speed, the parser and grammar used sacrifice functionalism (in the framework of dependency theory) in favour of projectivity. We have found that the resulting dependency structures, although cumbersome to handle, can be used by ExtrAns to find the syntactic and semantic dependencies needed in several of the linguistic processing stages. In particular, we focus on the minimal logical form generation.

In this paper we argue that questionanswering (QA) over technical domains is distinctly different from TREC-based QA or Web-based QA and it cannot benefit from data-intensive approaches. Technical questions arise in situations where concrete problems require specific answers and explanations. Finding a justification of the answer in the context of the document is essential if we have to solve a real-world problem. We show that NLP techniques can be used successfully in technical domains for high-precision access to information stored in documents. We present Extr-Ans, an answer extraction system over technical domains, its architecture, its use of logical forms for answer extractions and how terminology extraction becomes an important part of the system.

We present a deep-linguistic, practical system for the discovery of protein interactions in biomedical scientific literature. Our approach is based on probabilistic deep-syntactic parsing for dependency relations and a complex pattern search over the dependency trees to normalize alternations, diathesis, idiomatic expressions, and allowing semantic restrictions. We show that if reliable term recognition is available, performance on parsing biomedical texts is as accurate as on general text and present a successful Machine-Learning adaptation to the specific genre. A system for editing and testing KB queries formulated as lexico-syntactic patterns is presented. We conclude with an evaluation of the linguistic performance of the parser, a user-based practical evaluation and its evaluation in recent text mining competition.

The idea of the computer system capable of simulating understanding with respect to reading a story or passage and answering questions pertaining to it, has received increased attention within the NLP community as a means to develop and evaluate robust question answering methods. This research is concerned with the problem of generating an automated answer in the context of sophisticated knowledge representation, reasoning, and logical inferential processing. The research focused on Wh types of question to a restricted domain. External knowledge sources i.e. world knowledge (WK) and hypernyms matching procedure (HMP) were introduced. World knowledge led to the refinement of the ability of the system to extract the relevant answers. It provided a solution to the outstanding problem related to the ambiguity enclosed by anaphora and synonym words. Meanwhile, hypernyms refer to a more general or broad keywords, which was able to widen the results of the search. Thus, hypernyms matching procedure was to give a more coherent meaning of words, and eased in the process of extracting the answer based on a given question. This research found that the combined external knowledge sources on word dependencies improved the accuracy of the question answering system with respect to human performance measures.

Overview This paper introduces a new system that has been developed specifically for processing abbreviated text from an information-extraction point of view. Each of the three principal components of the system-text preprocessing, parsing for content, and discourse processing-is discussed in turn. Approaches used in the development of the knowledge sources are mentioned, with a particular focus on the linguistic issues involved. Examples of how sentences are processed are given, and strengths of the system are stressed. Finally, conclusions are drawn and future work and applications are sketched. Introducing LG-Soar The new system, called LG-Soar, represents the integration of three major processing components: (i) regular-expression-based text preprocessing; (ii) the Link Grammar parser; and (iii) the Soar intelligent agent architecture. The result is a robust, versatile text processing engine useful for difficult-to-handle input.  Why a new Soar parser? NL-Soar is:  Designed for cognitive modeling of natural language use  Not (yet) versatile enough to handle grammatically problematic text The project presented several interesting challenges from an NLP perspective. The overall goal was to mine content from problematic text. Most currently existing systems only perform well on well-structured, completely grammatical text. Another problem was to address complicated linguistic issues in the development of a usable system. Another goal was to output the information into a variety of usable formats. Finally, the project was meant to test the feasibility of integrating this particular set of components within a unified agent architecture. The system LG-Soar component of the system operates as follows: 1) An entry is read in from a pre-processed input file. 2) Each entry is split into individual sentences. 3) Each sentence is parsed with the Link Grammar. 4) The discourse representation module creates semantic/discourse representations of link content for all sentences in the entry. 5) Output is generated according to various formats. These steps are discussed in further detail in the rest of the paper.

Nowadays, due to the great advances in communication and storage media, there is more information available than ever before. This information can satisfy almost every information need; nevertheless, without the appropriate manage facilities, all of it is practically useless. This fact has motivated the emergence of several text processing applications that help in accessing large document collections. Currently, there are three main approaches for this purpose: information retrieval, information extraction, and question answering. Question answering (QA) systems aim to identify the exact answer to a question from a given document collection. This paper presents a survey of the Mexican experience in Spanish QA. In particular, it presents an overview of the participations of the Language Technologies Laboratory of INAOE (LabTL) in the Spanish QA evaluation task at CLEF, from 2004 to 2007. Through these participations, the LabTL has mainly explored two different approaches for QA: a language independent approach based on statistical methods, and a language dependent approach supported by sophisticated linguistic analyses of texts. It is important to point out that, due to these works, the LabTL has become one of the leading research groups in Spanish QA.

We describe our experience in developing a discourse-annotated corpus for community-wide use. Working in the framework of Rhetorical Structure Theory, we were able to create a large annotated resource with very high consistency, using a well-defined methodology and protocol. This resource is made publicly available through the Linguistic Data Consortium to enable researchers to develop empirically grounded, discoursespecific applications.

This paper presents two proposals based on semantic information, semantic roles and WordNet, for the answer extraction module of a general open-domain question answering (QA) system. The main objective of this research is to determine how the system performance is influenced by using this kind of information, and compare it with that of current QA systems based on named entities (NEs). NE-based QA systems achieve good results with NE-based questions. However, with common noun (CN) based questions, like ''Where is the stomach? In the abdomen", they fail, and this is the main reason for our study. In this paper our new proposals for answering different types of questions are evaluated and compared with an NE-based approach for both NE-based and CN-based questions. From the results obtained it may be concluded that, with the aid of our proposals, the QA system performs much better with CN-based questions when semantic information is used (semantic information F b¼1 ¼ 74:73% vs. NE F b¼1 ¼ 12:19%). Moreover, the more semantic information the system uses, the better the precision and correctness of the answer it achieves.

This paper describes the work done by Priberam in the development of a question answering (QA) system for Portuguese. The system was built using the company's NLP workbench and information retrieval technology. Special focus is given to the question analysis, document and sentence retrieval, and answer extraction stages. The paper discusses the system's performance in the context of the QA@CLEF 2005 evaluation.

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

This paper describes the work done by Priberam in the development of a question answering (QA) system for Portuguese. The system was built using the company's NLP workbench and information retrieval technology. Special focus is given to the question analysis, document and sentence retrieval, and answer extraction stages. The paper discusses the system's performance in the context of the QA@CLEF 2005 evaluation.

In this paper, we present a novel Biomedical Question Answering system, BioAMA: "Biomedical Ask Me Anything" on task 5b of the annual BioASQ challenge (Balikas et al., 2015). We focus on a wide variety of question types including factoid, list based, summary and yes/no type questions that generate both exact and wellformed 'ideal' answers. For summarytype questions, we combine effective IRbased techniques for retrieval and diversification of relevant snippets for a question to create an end-to-end system which achieves a ROUGE-2 score of 0.72 and a ROUGE-SU4 score of 0.71 on ideal answer questions (7% improvement over the previous best model). Additionally, we propose a novel Natural Language Inference (NLI) based framework to answer the yes/no questions. To train the NLI model, we also devise a transfer-learning technique by cross-domain projection of word embeddings. Finally, we present a two-stage approach to address the factoid and list type questions by first generating a candidate set using NER taggers and ranking them using both supervised and unsupervised techniques.

A typical knowledge-based question answering (KB-QA) system faces two challenges: one is to transform natural language questions into their meaning representations (MRs); the other is to retrieve answers from knowledge bases (KBs) using generated MRs. Unlike previous methods which treat them in a cascaded manner, we present a translation-based approach to solve these two tasks in one unified framework. We translate questions to answers based on CYK parsing. Answers as translations of the span covered by each CYK cell are obtained by a question translation method, which first generates formal triple queries as MRs for the span based on question patterns and relation expressions, and then retrieves answers from a given KB based on triple queries generated. A linear model is defined over derivations, and minimum error rate training is used to tune feature weights based on a set of question-answer pairs. Compared to a KB-QA system using a state-of-the-art semantic parser, our method achieves better results.

The availability of robust and deep syntactic parsing can improve the performance of all modules of a Question Answering system. In this article, this is illustrated using examples from our QA system Joost, a Dutch QA system which has been used for both open and closed domain QA. The system can make use of information found in the fully parsed version of the document collections. We demonstrate that this improves the performance of various components of the system, such as answer extraction and selection, lexical acquisition, off-line relation extraction, and passage retrieval. RÉSUMÉ. Une analyse syntaxique profonde et robuste améliore la performance d'un système de question-réponse. Dans cet article, nous le montrerons en donnant des exemples de notre système QR, appelé Joost. C'est un système néerlandais, qui a été appliqué au domaine général ainsi qu'au domaine restreint. Le système utilise l'information contenue dans une version analysée syntaxiquement du corpus des documents. Nous montrerons que l'utilisation de l'information syntaxique améliore certains modules de Joost, comme l'extraction et l'ordonnancement final des réponses, l'acquisition automatique d'information lexicale, l'extraction de faits hors ligne et la recherche de passages.

We investigate to what extent it helps to use an automatically generated instance list for resolving anaphoric definite NPs within the framework of an off-line answer extraction system. We compare three extraction methods. First, we looked at the extraction method that uses no anaphora resolution. Secondly, we determined the precision of the facts acquired by using instances for anaphora resolution. And thirdly, as a baseline we extracted facts by selecting the most recent named entity as the antecedent of the definite NP. From our experiments we can conclude that anaphora resolution using instances yields a high precision. Overall this precision is only 1% lower than the precision of the original answer extraction method without applying anaphora resolution. We outperformed the baseline of which the precision was not even half as high as the other two extraction methods. 1 We use the CLEF-corpus for our experiments. This corpus consists of newspaper text from 1994 and 1995.

This paper describes the participation of the Technical University of Catalonia in the CLEF 2009 Question Answering on Speech Transcripts track. We have participated in the English and Spanish scenarios of QAst. For both manual and automatic transcripts we have used a robust factual Question Answering that uses minimal syntactic information. We have also developed a NERC designed to handle automatic transcripts. We perform a detailed analysis of our results and draw conclusions relating QA performance to word error rate and the difference between written and spoken questions.

In this article, we present a factoid question-answering system, Sibyl, specifically tailored for question answering (QA) on spoken-word documents. This work explores, for the first time, which techniques can be robustly adapted from the usual QA on written documents to the more difficult spoken document scenario. More specifically, we study new information retrieval (IR) techniques designed or speech, and utilize several levels of linguistic information for the speech-based QA task. These include named-entity detection with phonetic information, syntactic parsing applied to speech transcripts, and the use of coreference resolution. Sibyl is largely based on supervised machine-learning techniques, with special focus on the answer extraction step, and makes little use of handcrafted knowledge. Consequently, it should be easily adaptable to other domains and languages. Sibyl and all its modules are extensively evaluated on the European Parliament Plenary Sessions English corpus, compa...

Petunjuk 1. Bacalah LKPD ini dengan seksama. 2. Diskusikan dan tuliskan jawaban yang tepat pada setiap pertanyaan yang disediakan bersama anggota kelompok masing-masing. 3. Tanyakan hal-hal yang tidak dimengerti kepada guru untuk mendapatkan penjelasan lebih lanjut.

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

QA is a application or specific type of information Retrieval. Today’s world of the web, information is store with large repositories of information. The Big Problem we face is that large amount of information available that allows us to find what is relevant and cannot be managed without automatic Search. We solve this problem by using Question answering System(QAS). The main goal of QA system is able to answer users Question in Natural Language. The QA system can fulfil the needs of user as they provide with faster and appropriate answers to user question.

We propose a hybrid architecture for the NTCIR-5 CLQA CC (Cross Language Question Answering from Chinese to Chinese) Task. Our system, the Academia Sinica Question-Answering System (ASQA), outputs exact answers to six types of factoid question: personal names, location names, organization names, artifacts, times, and numbers. The architecture of ASQA comprises four main components: Question Processing, Passage Retrieval, Answer Extraction, and Answer Ranking. ASQA successfully combines machine learning and knowledge-based approaches to answer Chinese factoid questions, achieving 37.5% and 44.5% Top1 accuracy for correct, and correct+unsupported answers, respectively.

The discovery of new chemical compounds is perceived as a key driver of the chemistry industry and many other economic sectors. The information about the new discoveries are usually disclosed in scientific literature and in particular, in chemical patents, since patents are often the first venues where the new chemical compounds are publicized. Despite the significance of the information provided in chemical patents, extracting the information from patents is costly due to the large volume of existing patents and its drastic expansion rate. The Cheminformatics Elsevier Melbourne University (ChEMU) evaluation lab 2020, part of the Conference and Labs of the Evaluation Forum 2020 (CLEF2020), provides a platform to advance the state-of-the-arts in automatic information extraction systems over chemical patents. In particular, we focus on extracting synthesis process of new chemical compounds from chemical patents. Using the ChEMU corpus of 1500 “snippets” (text segments) sampled from 17...

Nowadays, due to the great advances in communication and storage media, there is more information available than ever before. This information can satisfy almost every information need; nevertheless, without the appropriate manage facilities, all of it is practically useless. This fact has motivated the emergence of several text processing applications that help in accessing large document collections. Currently, there are three main approaches for this purpose: information retrieval, information extraction, and question answering. Question answering (QA) systems aim to identify the exact answer to a question from a given document collection. This paper presents a survey of the Mexican experience in Spanish QA. In particular, it presents an overview of the participations of the Language Technologies Laboratory of INAOE (LabTL) in the Spanish QA evaluation task at CLEF, from 2004 to 2007. Through these participations, the LabTL has mainly explored two different approaches for QA: a language independent approach based on statistical methods, and a language dependent approach supported by sophisticated linguistic analyses of texts. It is important to point out that, due to these works, the LabTL has become one of the leading research groups in Spanish QA.

This paper describes the development of a Question Answering (QA) system and its evaluation results in the Romanian-English cross-lingual track organized as part of the CLEF 1 2006 campaign. The development stages of the cross-lingual Question Answering system are described incrementally throughout the paper, at the same time pinpointing the problems that occurred and the way they were addressed. The system adheres to the classical architecture for QA systems, debuting with question processing followed, after term translation, by information retrieval and answer extraction. Besides the common QA difficulties, the track posed some specific problems, such as the lack of a reliable translation engine from Romanian into English, and the need to evaluate each module individually for a better insight into the system's failures.

The Question Interpretation module of QA@L 2 F, the question-answering system from L 2 F/INESC-ID, is thoroughly described in this paper, as well as the frame formalism 1 it employs. Moreover, the anaphora resolution process introduced this year, based on frames manipulation, is detailed. The overall results QA@L 2 F achieved at the CLEF competition and a brief overview on the system's evolution throughout the 2 years of joint evaluation are presented. The results of an evaluation to the QI module alone are also detailed here.

Today, language understanding systems do quite many useful things with processing natural language, even they are able to process the data much faster than humans are. Nevertheless, they do not have the same logical understanding of natural language yet as humans have and the interpretation capabilities of a language understanding system depending on the semantic theory is not sufficient in all aspects. The research is centered on some of the important issues that arise using it in natural language processing.

This paper describes the experiments of the TALP-UPC group for factoid and 'other' (definitional) questions at TREC 2005 Main Question Answering (QA) task. Our current approach for factoid questions is based on a voting scheme among three QA systems: TALP-QA (our previous QA system), Sibyl (a new QA system developed at DAMA-UPC and TALP-UPC), and Aranea (a web-based data-driven approach). For defitional questions, we used two different systems: the TALP-QA Definitional system and LCSUM (a Summarization-based system). Our results for factoid questions indicate that the voting strategy improves the accuracy from 7.5% to 17.1%. While these numbers are low (due to technical problems in the Answer Extraction phase of TALP-QA system) they indicate that voting is a succesful approach for performance boosting of QA systems. The answer to definitional questions is produced by selecting phrases using set of patterns associated with definitions. Its results are 17.2% of F-score in the best configuration of TALP-QA Definitional system.

Question Answer System (QAS) has manymethods in determining candidate answer and must have theright answer for each question. Previous QAS using Fuzzylogic focused on candidate and ranking of answer. However,the QAS needs improvement in determining the relevantanswer from the question and generating the correct answer inthe question answer process. In this paper, we propose a newmethod to combine fuzzy logic and retrieval passages to obtaina collection of relevant answers in order to obtain high answeraccuracy. We take relevant answers from the collection of theanswer document and choose the exact answer using fuzzylogic. Methods for the QAS are preprocessing, questionanalyzer, passage retrieval, passage scoring, scoring for similartext, measuring keyword and candidate answers, fuzzy logiccontroller, rules, and extraction answer. This study producedsignificantly relevant answers compared to the TF-IDFmethod. The performance of the system has improved theaccuracy of QAS by 80% which ...