Data Analysis for Emotion Identification in Text

This thesis investigates how to identify emotional sentences in an article using data analysis technologies. Two types of methods are proposed to solve the problem. A straightforward method of identifying emotional sentences is to formulate it as a classification problem. A classifier based on Linear Discriminant Analysis (LDA) is proposed for the classification on imbalanced data sets. The classifier fusion is also investigated to further improve the system performance by combining different classifiers and features. Emotion identification in text is formulated as a ranking problem that calculates the score of emotion which is hidden in every sentence. The sentences with higher emotion scores are predicted as emotional ones. The associations between objects should be updated if new objects are appended to a data set. The incremental learning of data association is discussed to make association based methods be able to adapt to new data.

Author dc:creator

• ZHANG ZHENGCHEN

• Emotion Identification, Data Analysis, Imbalanced Classification, Data Association, Ranking, Document Summarization