Exploring contextual factors from consumer reviews affecting movie sales: an opinion mining approach
- PDF / 1,447,889 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 21 Downloads / 233 Views
Exploring contextual factors from consumer reviews affecting movie sales: an opinion mining approach Li‑Chen Cheng1 · Chi‑Lun Huang1
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract In the age of Web 2.0, the rapid growth of user-generated content (e.g., consumer reviews) on the Internet offers ample avenues to search for information useful to both people and companies. Prior works in this field relating to movies have focused on the average rating and the number of comments. In this study, we used the content of consumer reviews and propose a novel framework integrating opinion mining and machine learning techniques to explore contextual factors influencing box-office revenue. Moreover, we analyzed movie review data from the website Internet Movie Database to examine the relationship among time periods, users’ opinion, and changes in box-office patterns. Experimental evaluations demonstrated that changes in different aspects of opinions effected a change in box-office revenue. Thus, movie marketers should monitor changes in the various aspects of online reviews and accordingly devise e-marketing strategies. Keywords Opinion mining · Sentiment analysis · Temporal abstraction · Consumer reviews
1 Introduction The movie industry is a highly competitive multibillion-dollar business [16]. In 2016, the global box-office revenue for all films worldwide reached $38.6 billion, up 1% from 2015. Movie box-office revenues indicate the financial performance of movies. However, the financial success of a movie is largely uncertain [16]. Typically, movie firms make large advertising investments to promote a movie prior to its release. In recent years, a low-budget movies, such as Cloverfield and The Dark Knight, have achieved considerable success by using viral online word-of-mouth (WOM) marketing [17]. Because movies belong to the experience * Li‑Chen Cheng [email protected] 1
Department of Computer Science and Information Management, Soochow University, Taipei 100, Taiwan, ROC
13
Vol.:(0123456789)
L.-C. Cheng, C.-L. Huang
goods category, people tend to share their feelings immediately after watching films and less as time goes by. WOM has a strong influence on consumers’ movie selection [16, 20]. Numerous online review websites are available on the Internet, and they are rapidly gaining in popularity. Consumers are encouraged to share their experiences and express their opinions on several well-known websites, such as Internet Movie Database (IMDb) and Amazon.com. Online reviews as a form of user-generated content (UGC) is increasingly becoming a major source of information to consumers in their search for, evaluation of, and choice of products [8, 14]. Online reviews influence consumers’ purchase decision-making process [4, 12, 30]. Thus, online reviews are a potentially valuable tool for firms and can be used to monitor consumer attitudes toward their products in real time; firms can then accordingly adapt their manufacturing, distribution, and marketing strategies to their a
Data Loading...
