Title: Identifying the causal relationship between social media content of a Bollywood movie and its box-office success - a text mining approach
Authors: Biplab Bhattacharjee; Amulyashree Sridhar; Anirban Dutta
Addresses: School of Management Studies, National Institute of Technology, Calicut, India ' Department of Biotechnology, PES Institute of Technology, BSK 3rd Stage, Bangalore, India ' School of Management, National Institute of Technology, Agartala, India
Abstract: Movie marketing strategies have undergone a rapid metamorphosis over the years with the progress in technological innovations and advent of social media. Social media gives a two way interacting platform and such interactions generate voluminous textual content which can be a source for deriving new insights into the customer behavioural dynamics and can also act as a handy tool for revenue enhancement. This study is designed to understand whether the polarity of the social media content of Bollywood movies can essentially reveal any insights about the potential box office revenues. The initial steps involved data collection from social media, followed by text mining to identify the sentiments about a movie. Furthermore, the relationship between the sentiments captured from social media and total revenue generated was explored in both pre-release and post-release scenarios and linear regression models were built. The model can be further improved by incorporating additional metrics.
Keywords: social media content; sentiment analysis; Bollywood movies; Bollywood films; box-office success; text mining; movie marketing; film marketing; customer behaviour; behavioural dynamics; revenue enhancement; box office revenues.
International Journal of Business Information Systems, 2017 Vol.24 No.3, pp.344 - 368
Received: 15 Jun 2015
Accepted: 17 Aug 2015
Published online: 06 Feb 2017 *