Abstract
Multinomial naive Bayes classifies have been widely used for the probabilistic text classification. However, their parameter estimation method sometimes generates inappropriate probabilities. In this paper, we propose a topic document model approach for naive Bayes text classification, where their parameters are estimated with an expectation from the training documents. Experiments are conducted on Reuters 21578 and 20 Newsgroup collection, and our proposed approach obtained a significant improvement in performance over the conventional approach.
Original language | English |
---|---|
Pages (from-to) | 391-392 |
Number of pages | 2 |
Journal | SIGIR Forum (ACM Special Interest Group on Information Retrieval) |
DOIs | |
Publication status | Published - 2002 |
Event | Proceedings of the Twenty-Fifth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval - Tampere, Finland Duration: 2002 Aug 11 → 2002 Aug 15 |
ASJC Scopus subject areas
- Management Information Systems
- Hardware and Architecture