sis 18(19): e1

Research Article

Deep Level Markov Chain Model for Semantic Document Retrieval

Download1179 downloads
Cite
BibTeX Plain Text
  • @ARTICLE{10.4108/eai.19-6-2018.155443,
    author={Linh Bui Khanh and Ha Nguyen Thi Thu and Tinh Dao Thanh},
    title={Deep Level Markov Chain Model for Semantic Document Retrieval},
    journal={EAI Endorsed Transactions on Scalable Information Systems},
    volume={5},
    number={19},
    publisher={EAI},
    journal_a={SIS},
    year={2018},
    month={9},
    keywords={Big data, information retrieval, feature reduction, Markov chain, probability inference},
    doi={10.4108/eai.19-6-2018.155443}
}
  • Linh Bui Khanh
    Ha Nguyen Thi Thu
    Tinh Dao Thanh
    Year: 2018
    Deep Level Markov Chain Model for Semantic Document Retrieval
    SIS
    EAI
    DOI: 10.4108/eai.19-6-2018.155443
Linh Bui Khanh1, Ha Nguyen Thi Thu1,*, Tinh Dao Thanh2
  • 1: Electric Power University, 235 Hoang Quoc Viet, Hanoi, Vietnam
  • 2: Le Qui Don Technical University, 236 Hoang Quoc Viet, Hanoi, Vietnam
*Contact email: hantt@epu.edu.vn

Abstract

The task of researching and developing information retrieval systems is becoming important in the big data age. Current search methods try to mention to fast searching based on keyword matching or similar semantic between query and documents but have not got a really effective engine for semantic search . In this paper, we propose a method for information retrieval based on probability inference with the DLMC model to search by semantic equivalents and a topic word with score for fast searching. Results of the experimental with 952 Vietnamese documents show that our method is really effective for Vietnamese document retrieval system.

Keywords
Big data, information retrieval, feature reduction, Markov chain, probability inference
Received
2018-04-29
Accepted
2018-07-18
Published
2018-09-10
Publisher
EAI
http://dx.doi.org/10.4108/eai.19-6-2018.155443

Copyright © 2018 Linh Bui Khanh et al., licensed to EAI. This is an open access article distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/3.0/), which permits unlimited use, distribution and reproduction in any medium so long as the original work is properly cited.