Home > Archive > 2017 > Volume 7 Number 4 (Aug. 2017) >
IJMLC 2017 Vol.7(4): 72-75 ISSN: 2010-3700
DOI: 10.18178/ijmlc.2017.7.4.623

Box Office Revenue Prediction Using Dual Sentiment Analysis

Prashant Rajput, Priyanka Sapkal, and Shefali Sinha

Abstract—Twitter is amongst the most widely used social networking website and it is also a reliable source of mass opinion. Success of a movie can be predicted by analyzing tweets and examining the impact of movie on the mob. Pre-release buzz can also be captured through tweets. This knowledge helps in predicting the success of a movie and its approximate revenue. In this paper, Dual Sentiment Analysis (DSA) is used for sentiment analysis of tweets that avoids sentiment classification problems and improves performance. Along with sentiment analysis of tweets, contribution of other factors such as star cast, holiday effect, sequel and genre are also considered. Finally, multivariate linear regression is performed on all above-mentioned factors to predict the Box Office revenue of a movie. The results show that this proposed system performs better while providing better accuracy.

Index Terms—Natural language processing, sentiment analysis, opinion mining, machine learning, social media.

Prashant Rajput is with Computer Science Department, University of California, Los Angeles (UCLA), United States (e-mail: prashanthrajput@ucla.edu).
Priyanka Sapkal is with Persistent Systems, India (e-mail: priyanka_sapkal@persistent.com).
Shefali Sinha is with State Bank of India (SBI), India (e-mail: shefali.sinha@sbi.co.in).

[PDF]

Cite: Prashant Rajput, Priyanka Sapkal, and Shefali Sinha, "Box Office Revenue Prediction Using Dual Sentiment Analysis," International Journal of Machine Learning and Computing vol. 7, no. 4, pp. 72-75, 2017.

General Information

  • ISSN: 2010-3700 (Online)
  • Abbreviated Title: Int. J. Mach. Learn. Comput.
  • Frequency: Bimonthly
  • DOI: 10.18178/IJMLC
  • Editor-in-Chief: Dr. Lin Huang
  • Executive Editor:  Ms. Cherry L. Chen
  • Abstracing/Indexing: Inspec (IET), Google Scholar, Crossref, ProQuest, Electronic Journals Library.
  • E-mail: ijmlc@ejournal.net


Article Metrics