eduzhai > Applied Sciences > Engineering >

Investigating People’s Sentiment from Twitter Data for Smart Cities: A Survey

  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: The usage of World Wide Web (WWW) and speed of internet has facilitated information consumers to become information producers via social media. Rapid growths of social media in recent years instigate to realize city officials that it can be a great platform to understand their citizens. To provide better services to citizens, the authorities are trying to move closer to them. Twitter sentiment analysis can be a one of the powerful tools to analyze valuable reflection of public perception. This paper is focused on Twitter sentiment analysis for smart cities. Effective feature engineering of twitter text is one of the key aspects for Twitter Sentiment Analysis along with different analysis approaches. These variables are analyzed and some related work is included for discussion. Several characteristics of tweets such as the size of tweets and resulting compact, novel language with Twitter specific communication elements, a strong sentiment class imbalance, and stream-based tweet generation complicate the analysis and challenge Twitter Sentiment Analysis approaches, resulting in generally poor sentiment classification performance. This paper also review methods devised to address these challenges.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×