Reliability of Sentiment Mining Tools: A comparison of Semantria and Social Mention

Abstract

Social Media platforms have become quintessential for user-generated content and consumer opinions. As a result, vast amounts of commercially and freely available sentiment mining tools have emerged, however it remains unclear how reliable these tools are. In this paper, I evaluate the reliability as well as the features of the sentiment mining tools Semantria and Social Mention. This study is a sentiment analysis of 12 different car models that were applied to both sentiment mining tools. In addition to presenting a comparison of outputs obtained from two different sentiment mining tools, an analysis was conducted that compares the sentiment and passion outputs for three social media platforms. The results show significant differences in outputs for sentiment mining tools. Furthermore, statistically signifcant as well as observational differences were also apparent for the outputs from three social media platforms. The results have theoretical as well practical implications for different groups, including academics and practitioners using sentiment mining tools as input for future research and decision-making.

Cite this paper

@phdthesis{lawrence2014reliability,
  title = {{Reliability of Sentiment Mining Tools: A comparison of Semantria and Social Mention}},
  author = {Lawrence, Linus Philip},
  year = {2014}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.

Related papers