Information Extraction System Based on Hidden Markov Model

Authors:
Dong-Chul Park;Vu Thi Huong;Dong-Min Woo;Duong Ngoc Hieu;Sai Thi Ninh
Affiliations:
Dept. of Information Engineering, Myong Ji University, Korea;Dept. of Information Engineering, Myong Ji University, Korea;Dept. of Information Engineering, Myong Ji University, Korea;Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology, Vietnam;Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology, Vietnam
Venue:
ISNN '09 Proceedings of the 6th International Symposium on Neural Networks on Advances in Neural Networks
Year:
2009

Citing 6
Cited 0

A critical investigation of recall and precision as measures of retrieval system performance

ACM Transactions on Information Systems (TOIS)
Learning Information Extraction Rules for Semi-Structured and Free Text

Machine Learning - Special issue on natural language learning
Learning dictionaries for information extraction by multi-level bootstrapping

AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
IEPAD: information extraction based on pattern discovery

Proceedings of the 10th international conference on World Wide Web
A Fully Automated Object Extraction System for the World Wide Web

ICDCS '01 Proceedings of the The 21st International Conference on Distributed Computing Systems
PEWeb: Product Extraction from the Web Based on Entropy Estimation

WI '04 Proceedings of the 2004 IEEE/WIC/ACM International Conference on Web Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

A novel approach using Hidden Markov Model (HMM) for the task of finding prices of products on internet sites is proposed in this paper. The proposed Information Extraction System based on HMM (IESHMM) utilizes HMM for its capability to process temporal information. The proposed IESHMM first processes web pages that are returned from search engines and then extracts specific fields such as prices, descriptions, locations, images of products, and other information of interest. The proposed IESHMM is evaluated with real-world problems and compared with a conventional method. The results show that the proposed IESHMM outperforms the other method by 22.9 % and 37.2% in terms of average recall and average precision, respectively.