loading
Documents

Research.Publish.Connect.

Paper

Paper Unlock

Authors: Fumiyo Fukumoto and Yoshimi Suzuki

Affiliation: University of Yamanashi, Japan

ISBN: 978-989-758-158-8

ISSN: 2184-3228

Keyword(s): Feature Selection, Latent Dirichlet Allocation, Temporal-based Features, Text Categorization, Timeline Adaptation, Transfer Learning.

Related Ontology Subjects/Areas/Topics: Artificial Intelligence ; Clustering and Classification Methods ; Information Extraction ; Knowledge Discovery and Information Retrieval ; Knowledge-Based Systems ; Symbolic Systems

Abstract: This paper addresses text categorization problem that training data may derive from a different time period from the test data. We present a method for text categorization that minimizes the impact of temporal effects. Like much previous work on text categorization, we used feature selection. We selected two types of informative terms according to corpus statistics. One is temporal independent terms that are salient across full temporal range of training documents. Another is temporal dependent terms which are important for a specific time period. For the training documents represented by independent/dependent terms, we applied boosting based transfer learning to learn accurate model for timeline adaptation. The results using Japanese data showed that the method was comparable to the current state-of-the-art biased-SVM method, as the macro-averaged F-score obtained by our method was 0.688 and that of biased-SVM was 0.671. Moreover, we found that the method is effective, especially whe n the creation time period of the test data differs greatly from that of the training data. (More)

PDF ImageFull Text

Download
CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 3.236.112.98

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Fukumoto, F. and Suzuki, Y. (2015). Temporal-based Feature Selection and Transfer Learning for Text Categorization.In Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1 KDIR: KDIR, (IC3K 2015) ISBN 978-989-758-158-8, ISSN 2184-3228, pages 17-26. DOI: 10.5220/0005593100170026

@conference{kdir15,
author={Fumiyo Fukumoto. and Yoshimi Suzuki.},
title={Temporal-based Feature Selection and Transfer Learning for Text Categorization},
booktitle={Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1 KDIR: KDIR, (IC3K 2015)},
year={2015},
pages={17-26},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005593100170026},
isbn={978-989-758-158-8},
}

TY - CONF

JO - Proceedings of the 7th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 1 KDIR: KDIR, (IC3K 2015)
TI - Temporal-based Feature Selection and Transfer Learning for Text Categorization
SN - 978-989-758-158-8
AU - Fukumoto, F.
AU - Suzuki, Y.
PY - 2015
SP - 17
EP - 26
DO - 10.5220/0005593100170026

Login or register to post comments.

Comments on this Paper: Be the first to review this paper.