Regresar

Polarity analysis in different types of documents

Abstract:

Classically, Sentiment Analysis aims to identify the writer’s attitude toward individuals, events, or topics, but this paper proposes an inverse approach, a process for sentiment/polarity identification in documents. Particularly, our approach can be used to pbkp_redict the possible effect of a written text on the users. Thus, this paper presents the Sentiment Classification problem in texts, and proposes a strategy to classify their polarity (positive/negative). For this, the paper analyses three keywords extraction methods from the text, and defines a process for an automatic identification of its polarity. The features/keywords extracted are analyzed by the polarity analysis process, to determine the positive/negative connotation of the text. Several experiments have been carried out in different datasets, with different forms to consider the extracted keywords.