B-18-30 Sentiment Analysis in Twitter for Multiple Topics : How to Detect the Polarity of Tweets Regardless of Their Topic

• Bouazizi Mondher

  Graduate School of Science and Technology, Keio University

• Ohtsuki Tomoaki

  Department of Information and Computer Science, Faculty of Science and Technology, Keio University

抄録

Sentiment analysis refers to the extraction and aggregation of subjective information out of a big amount of data. It is used to summarize users' level of satisfaction towards commercial products or anticipating the results of elections, etc. However, sentiment analysis has new applications with the appearance of new businesses such as ads-based businesses. Studying the passions of users and displaying ads based on what interests them cannot be done using the conventional approaches where a set of texts dealing with one object is analyzed and an overall opinion is extracted. In this article we propose a new method for sentiment analysis that classifies tweets regardless of their topic. We provide a new set of features to be used by machine learning algorithms to classify texts belonging to different topics.

収録刊行物

• 電子情報通信学会総合大会講演論文集

  電子情報通信学会総合大会講演論文集 2015 (2), 577-, 2015-02-24

  一般社団法人電子情報通信学会

詳細情報

• CRID: 1570572702907364096
• NII論文ID: 110009928226
• NII書誌ID: AN10471452
• 本文言語コード: en
• データソース種別:
  • CiNii Articles