Department of Computer Science, College of Computer Science and Engineering, Taibah University, Medina, KSA.

---

Abstract— Sentiment analysis is a characteristic task that aimsto detectthe sentiment of opinions incontent.Twitter sentiment analysis (TSA) is a promising field that has gained attentionin the last decade. Investigators in the TSA field have faced difficultiescomparing existing TSA techniques, as there is no agreed systematic framework. This meansthat the evaluation of existing techniques relies on selecting different datasets without meaningful justification. Another issue that arises whencomparing different TSA techniques is that there are no unified metrics.Some researchers select classification accuracy and otherschoose recall, precision, and F-measure metrics. In this paper, we proposea framework called Evaluation Framework for Twitter Sentiment Analysis (EFTSA)for TSA evaluation based on individual or multiple datasets. This would help researchers comparetheir Twitter sentiment approaches against others

Index Terms— Sentiment analysis, twitter, evaluation, metrics..

[PDF]

Cite: Abdullah Alsaeedi, "EFTSA: Evaluation Framework for Twitter Sentiment Analysis," Journal of Software vol. 14, no. 1, pp. 页码, 2019.