AISQA - An Artificial Immune Question Answering System

Full Text (PDF, 366KB), PP.28-34

Views: 0 Downloads: 0

Author(s)

Mohsen Shakiba Fakhr 1,* Mohammad Saniee Abadeh 2

1. Department of Computer Engineering, dezful branch, Islamic Azad University, Dezful, Iran

2. Electrical and Computer Engineering College, Tarbiat Modares University, Tehran, Iran

* Corresponding author.

DOI: https://doi.org/10.5815/ijmecs.2012.03.04

Received: 6 Dec. 2011 / Revised: 10 Jan. 2012 / Accepted: 15 Feb. 2012 / Published: 8 Mar. 2012

Index Terms

QA, GA, Artificial Immune System, Mutation

Abstract

Question answering (QA) is the task of automatically answering a question posed in natural language. At this time, there exists several QA approaches, and, according to recent evaluation results, most of them are complementary. Some of them use the evolutionary algorithms, such as the genetic algorithm, in itself. In this paper we propose a question answering system that uses the artificial immune algorithms, for searching in the knowledge base to find the right answer. This algorithm is one of the evolutionary algorithms. Search is based on two features: (i) the compatibility between question and answer types, (ii) the overlap and non-overlap information between the question-answer pair. Experimental results are encouraging; they indicate significant increases in the accuracy of proposed system, in comparison with the previous systems.

Cite This Paper

Mohsen Shakiba Fakhr, Mohammad Saniee Abadeh, "AISQA - An Artificial Immune Question Answering System", International Journal of Modern Education and Computer Science (IJMECS), vol.4, no.3, pp.28-34, 2012. DOI: 10.5815/ijmecs.2012.03.04

Reference

[1]M. Shamsfard, M. Arab Yarmohammadi, " A Semantic Approach to Extract the Final Answer in SBUQA Question Answering System", International Journal of Digital Content Technology and its Applications, Volume 4, Number 7, 2010.
[2]X. Li, D. Roth, "Learning question classifiers", In COLING 2002, The 19th International Conference on Computational Linguistics, pp. 556–562, 2002.
[3]A. Ghobadi-Tapeh, M. Rahgozar, "A knowledge-based question answering system for B2C eCommerce", Elsevier Knowledge-Based Systems 21-journal homepage:www.elsevier.com/locate/knosys, PP. 946- 950, 2008.
[4]http://en.wikipedia.org/wiki/Question_answerig
[5]H. Baayen, et al., "Advances in Open Domain Question Answering", Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands, 2008.
[6]M.A. Yarmohamadi, "Answer Extraction from retrieved documents in a question answering system", MS Thesis, Computer Engineering Department, Faculty of Electrical and Computer Engineering, Shahid Beheshti University, Tehran, Iran, 2007.
[7]X. Lin, H. Liu, P. Lin, M. Wang, "Chinese Question Classification Using Alternating and Iterative One-against-One Algorithm" , Journal of convergence information technology, Volume 5, Number 3, May 2010.
[8]G. Salton, "The SMART Information Retrieval System", Prentice Hall, Englewood Cliffs, NJ, 1971.
[9]D. Dimmick, G. O'Brien, P. Over and W. Rogers, "Guide to Z39.50/Prise 2.0: Its Installation, Use, & Modification", Gaithersburg, Maryland, USA, 1998.
[10]A.G. Figueroa and G. Neumann,"Genetic Algorithms For Data-Driven Web Question Answering", journal, Evolutionary Computation, Volume 16 Issue 1, MIT Press Cambridge, 2008.