Multi-label text categorization forecasting probability problem using support vector machine techniques

Hui Min Chiang, Tai-Yue Wang, Yu Min Chiang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The pervasiveness of information available on the Internet means that increasing numbers of documents must be classified. Text categorization is not only undertaken by domain experts, but also by automatic text categorization systems. Therefore, a text categorization system with a multi-label classifier is necessary to process the large number of documents. In this study, a proposed multi-label text categorization system is developed to classify multi-label documents. Data mapping is performed to transform data from a high-dimensional space to a lower-dimensional space with paired SVM output values, thus lower the complexity of the computation. A pair-wise comparison approach is applied to set the membership function in each predicted class to judge all possible classified classes. Finally, the overlapped area of two classes is obtained from the decision function to determine where a document is classified. A comparative study is performed on multi-label approaches using Reuter's data sets. The results of the empirical experiment indicate that the proposed multi-label text categorization system performs better than other methods in terms of overall performance indices. Additionally, the probability of 0.5 for model membership function is a good criterion to judge between correctly and incorrectly classified documents from the results of the empirical experiment.

Original languageEnglish
Title of host publicationInformation Technologies in Environmental Engineering
Subtitle of host publicationNew Trends and Challenges, ITEE 2011
PublisherKluwer Academic Publishers
Pages39-48
Number of pages10
ISBN (Print)9783642195358
DOIs
Publication statusPublished - 2011 Jan 1
Event5th International Symposium on Information Technologies in Environmental Engineering, ITEE 2011 - Poznan, Poland
Duration: 2011 Jul 62011 Jul 8

Publication series

NameEnvironmental Science and Engineering (Subseries: Environmental Science)
ISSN (Print)1863-5520

Other

Other5th International Symposium on Information Technologies in Environmental Engineering, ITEE 2011
CountryPoland
CityPoznan
Period11-07-0611-07-08

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Environmental Engineering
  • Information Systems

Cite this

Chiang, H. M., Wang, T-Y., & Chiang, Y. M. (2011). Multi-label text categorization forecasting probability problem using support vector machine techniques. In Information Technologies in Environmental Engineering: New Trends and Challenges, ITEE 2011 (pp. 39-48). (Environmental Science and Engineering (Subseries: Environmental Science)). Kluwer Academic Publishers. https://doi.org/10.1007/978-3-642-19536-5_3