With the increased incidence of depressive disorders, many psychiatric websites have developed community-based services such as message boards, web forums and blogs for public access. Using machine learning approaches, we can identify user's emotions from such forum and blog posts to recognize the variance in depressive disorders automatically. The incremental association language feature is applied in this research to discover words with high information content in sentences. In past research, the overlap-category in building a feature has not been considered. Hence, this work makes a pioneering attempt to develop a model for emotion classification with overlap-category consideration. This research applies association rule mining to discover words appearing with high frequency in a sentence and to avoid a feature-overlap in categories simultaneously. The approach is named Association Language Features by Category (ALFC). The experimental results show that ALFC features have ability to distinguish between the various categories. The result has been compared with the approach of baseline and mutual information which use single words and correlation measures respectively.View full textDownload full textKeywordsassociation language features, emotion classification, feature-overlap, nature language processingRelated var addthis_config = { ui_cobrand: "Taylor & Francis Online", services_compact: "citeulike,netvibes,twitter,technorati,delicious,linkedin,facebook,stumbleupon,digg,google,more", pubid: "ra-4dff56cd6bb1830b" }; Add to shortlist Link Permalink http://dx.doi.org/10.1080/02533839.2011.591964
展开▼
