Filters








111 Hits in 14.2 sec

The LogAnswer Project at CLEF 2008: Towards Logic-Based Question Answering

Ingo Glöckner, Björn Pelzer
2008 Conference and Labs of the Evaluation Forum  
To this end, the optimized deductive subsystem is combined with shallow techniques by machine learning.  ...  The same background knowledge as in the MAVE validator of the IICS presented at CLEF 2007 was used: 10,000 lexical-semantic relations (e.g. describing nominalizations), 109 logical rules, and a list of  ...  While the combination of both provers worked well in earlier experiments based on cross-validation reported in [6] , results in the QA@CLEF 2008 task were slightly worse for the combined method compared  ... 
dblp:conf/clef/GlocknerP08a fatcat:dq5vlcgaxnbpzg3tdayskzfapi

Combining Logic and Machine Learning for Answering Questions [chapter]

Ingo Glöckner, Björn Pelzer
2009 Lecture Notes in Computer Science  
Robustness to gaps in the background knowledge and errors of linguistic analysis is achieved by combining the optimized deductive subsystem with shallow techniques by machine learning.  ...  Its main innovation is the use of logic for simultaneously extracting answer bindings and validating the corresponding answers.  ...  The paper explains the design of the LogAnswer prototype that tries to overcome these problems by combining logic and machine learning.  ... 
doi:10.1007/978-3-642-04447-2_47 fatcat:2huinbl3cbairivvjmg53pfity

Efficient Question Answering with Question Decomposition and Multiple Answer Streams [chapter]

Sven Hartrumpf, Ingo Glöckner, Johannes Leveling
2009 Lecture Notes in Computer Science  
The German question answering (QA) system IRSAW (formerly: InSicht) participated in QA@CLEF for the fifth time.  ...  The answer validator introduced in the previous year was replaced with the faster RAVE validator designed for logic-based answer validation under time constraints.  ...  Results may have been better with another machine translation service for QA@CLEF 2008.  ... 
doi:10.1007/978-3-642-04447-2_49 fatcat:6grdchnn7va4vl5k7zoo2ncsly

The Contribution of FaMAF at QA@CLEF 2008. Answer Validation Exercise

Julio J. Castillo
2008 Conference and Labs of the Evaluation Forum  
We use two different approaches using machine learning, specifically Support Vector Machine as classifier. The results show an increment over the baselines, however enhanced is needed.  ...  This shows an increment of 23.53% over the QA accuracy baseline.  ...  This measure is used for compare the AV systems with QA systems presented in QA@CLEF. Baselines for comparing AV systems performance with QA systems in English.  ... 
dblp:conf/clef/Castillo08 fatcat:u2n4oklfd5hjvlwwnzm4ofgaqy

Connecting Question Answering and Conversational Agents

Ulli Waltinger, Alexa Breuing, Ipke Wachsmuth
2012 Künstliche Intelligenz  
First, we present a comprehensive question classification experiment based on machine learning using two different datasets and various feature sets for the German language.  ...  Research results in the field of Question Answering (QA) have shown that the classification of natural language questions significantly contributes to the accuracy of the generated answers.  ...  By combining word features and machine learning-based Support Vector Machines (SVM ), they obtained an accuracy of 0.82 on English, 0.88 on Italian and 0.80 on Spanish.  ... 
doi:10.1007/s13218-012-0208-1 fatcat:7nyicxslqzczvlyv4hdvuoumce

Survey on Answer Validation for Indonesian Question Answering System (IQAS)

Abdiansah Abdiansah, Azhari Azhari, Anny K. Sari
2018 International Journal of Intelligent Systems and Applications  
One of the important issues in IQAS is Answer Validation (AV), which is a system that can determine the correctness of QAS.  ...  Research on Question Answering System (QAS) has been done mainly in English.  ...  The validation module was based on a machine learning approach.  ... 
doi:10.5815/ijisa.2018.04.08 fatcat:r3r6tygg5nawzkgkwj7jh42ele

A Case Based Reasoning Approach for Answer Reranking in Question Answering [article]

Karl-Heinz Weis
2015 arXiv   pre-print
In this document I present an approach to answer validation and reranking for question answering (QA) systems.  ...  In the experiments based on QA@CLEF questions, the best learned models make heavy use of CBR features.  ...  Experimental results The data set used for the experiments was generated by retrieving answer candidates for questions in the QA@CLEF 2007 and QA@CLEF 2008 test sets for German. 8 Due to the focus of CBR  ... 
arXiv:1503.02917v1 fatcat:tenrbjmd2zeu3ghxcrxtpreefq

Overview of the Answer Validation Exercise 2006 [chapter]

Anselmo Peñas, Álvaro Rodrigo, Valentín Sama, Felisa Verdejo
2007 Lecture Notes in Computer Science  
The Answer Validation Exercise at the Cross Language Evaluation Forum is aimed at developing systems able to decide whether the answer of a Question Answering system is correct or not.  ...  Validation setting. 9 groups have participated with 16 runs in 4 different languages.  ...  We are grateful to all the people involved in the organization of the QA track (specially to the coordinators at CELCT, Danilo Giampiccolo and Pamela Forner).  ... 
doi:10.1007/978-3-540-74999-8_32 fatcat:sge4gbrn6rhbxkhc4pi3lye3zu

An Integrated Machine Learning and Case-Based Reasoning Approach to Answer Validation

Ingo Glockner, Karl-Heinz Weis
2012 2012 11th International Conference on Machine Learning and Applications  
In our experiments on QA@CLEF questions, the best learned models make heavy use of CBR features.  ...  for validating answer candidates for new questions.  ...  Moreover, we aim at a joint framework that integrates CBR with proven techniques for answer validation (such as logical validation) by means of machine learning (ML).  ... 
doi:10.1109/icmla.2012.90 dblp:conf/icmla/GlocknerW12 fatcat:n5dbtsnv6fannorgtlc4hpio4m

Overview of the Answer Validation Exercise 2008 [chapter]

Álvaro Rodrigo, Anselmo Peñas, Felisa Verdejo
2009 Lecture Notes in Computer Science  
The Answer Validation Exercise at the Cross Language Evaluation Forum is aimed at developing systems able to decide whether the answer of a Question Answering system is correct or not.  ...  Validation setting. 9 groups have participated with 16 runs in 4 different languages.  ...  We are grateful to all the people involved in the organization of the QA track (specially to the coordinators at CELCT, Danilo Giampiccolo and Pamela Forner).  ... 
doi:10.1007/978-3-642-04447-2_35 fatcat:2u6savrl7fgbxl3tpolnrfvrdq

Learning to Translate for Multilingual Question Answering

Ferhan Ture, Elizabeth Boschee
2016 Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing  
We build a feature for each combination of translation direction and method, and train a model that learns optimal feature weights.  ...  In multilingual question answering, either the question needs to be translated into the document language, or vice versa.  ...  We would also like to thank the anonymous reviewers for their helpful feedback.  ... 
doi:10.18653/v1/d16-1055 dblp:conf/emnlp/TureB16 fatcat:eja3yohw4zaj3obhl2mgmfovsq

Learning to Translate for Multilingual Question Answering [article]

Ferhan Ture, Elizabeth Boschee
2016 arXiv   pre-print
We build a feature for each combination of translation direction and method, and train a model that learns optimal feature weights.  ...  In multilingual question answering, either the question needs to be translated into the document language, or vice versa.  ...  We would also like to thank the anonymous reviewers for their helpful feedback.  ... 
arXiv:1609.08210v1 fatcat:7rnfctatzbbytppuuxc5aizbdm

Extending a Logic-Based Question Answering System for Administrative Texts [chapter]

Ingo Glöckner, Björn Pelzer
2010 Lecture Notes in Computer Science  
LogAnswer is a question answering (QA) system for German that uses machine learning for integrating logic-based and shallow (lexical) validation features.  ...  The system was especially successful at detecting wrong answers, with 73% correct rejections.  ...  As to definition questions, the training set for learning the validation model (using QA@CLEF 2007/2008 questions) included too few examples for successful application of our ML technique.  ... 
doi:10.1007/978-3-642-15754-7_30 fatcat:dt336yocuve7zb2guoisbv7sjm

INAOE at AVE 2007: Experiments in Spanish Answer Validation

Alberto Téllez-Valero, Manuel Montes-y-Gómez, Luis Villaseñor Pineda
2007 Conference and Labs of the Evaluation Forum  
This paper describes the INAOE's answer validation system evaluated at the Spanish track of the AVE 2007.  ...  This system is based on a supervised learning approach that considers two kinds of attributes.  ...  We also like to thanks to the CLEF organizing committee as well as to the EFE agency for the resources provided.  ... 
dblp:conf/clef/Tellez-ValeroMV07 fatcat:3xztrb4sqvde7dlwaaqldvflnu

The LogAnswer Project at CLEF 2009

Ingo Glöckner, Björn Pelzer
2009 Conference and Labs of the Evaluation Forum  
LogAnswer uses a machine learning (ML) approach based on rank-optimizing decision trees for integrating logic-based and shallow (lexical) validation features.  ...  The LogAnswer system, a research prototype of a question answering (QA) system for German, participates in QA@CLEF for the second time.  ...  This is because the training set used for learning the validation model of LogAnswer contains annotated results of LogAnswer for the QA@CLEF 2007 and QA@CLEF 2008 questions.  ... 
dblp:conf/clef/GloecknerP09 fatcat:53rb5z5jxvh2nahnabdmfwusby
« Previous Showing results 1 — 15 out of 111 results