Hierarchical Classification in Text Mining for Sentiment Analysis
Jinyan Li, Simon James Fong, Yan Zhuang, Richard Khoury · 2014
Sentiment analysis in text mining is known to be a challenging task. Sentiment is subtly reflected by the tone, affective state or emotion of a writer's expression in words. Conventional text mining techniques which are based on keyword frequency counting usually run short of accurately detecting such subjective information implied in the text. In this paper we evaluated several popular classification algorithms, along with three filtering schemes. The filtering schemes progressively shrink the original dataset, with respect to the contextual polarity and frequent terms of a document. In general the proposed approach is coined hierarchical classification. The effects of the approach in different combination of classification algorithms and filtering schemes are discussed over three sets of controversial online news articles where binary and multi-class classifications are applied.