With the advent of Machines and Algorithms, humans have been exploring means to predict future events. From Nostradamus to Arthur C. Clarke who predicted global satellite communication in 1945, we have always been looking at predicting future events.
Data analysis can help detect miniscule spikes that can have a rippling effect on the business. This intelligence can help businesses predict and plan their activities to as close as real. Let us examine this with an example that gives you a better idea of how this works. Our analyst S. Aditya worked on a predictive model for movie success and sentiment analysis along with Mohan Raj, School of Business, Alliance University.
Their research objectives: Finding out how can we predict the success of a movie by measuring various variables of data this is available. Main research components.
- A. Variables that influence the frequency of movie watch
- B. Predicting the success of a movie by developing a model
- C. Sentiment analysis of selected two movies.
The U.S movie industry generated 564 billion USD in revenue by the end of 2014. It is predicted that the entertainment industry will grow to over 679 billion USD in value over the next four years. It is projected that 80% of the movie industry’s profits over the last decade is generated from just 6% of the films released; 78% of movies have lost money of the same time period.
Indian Movie Industry is different from its counterparts in many ways especially in terms of different movies in different languages and low ticket prices. In 2013, the Indian film industry generated 1.68 billion USD out of its total revenue of 2.07 billion from overseas and domestic box office collections According to the Deloitte and ASSOCHEM report, researchers predict that theatrical revenue in India will be increased from $1.78 billion to $2.13 billion by FY2017. That makes theatrical circuits account for some 74% of total income. The Indian scenario works a lot different than the western movies; a lot of importance is normally given to different parameters such as celebrity appeal, the movie album and others, which is an integral part of the movie itself; unlike, the western movies.
The study is descriptive in nature. The research design comprises of both qualitative and quantitative techniques. The factors that influence the frequency of watching movie are identified by literature review and quantitative approach is used to predict the success of a movie by developing a model. Multiple Regression analysis and Discriminant analysis are used for developing prediction model to predict the success of a movie. Sentiment analysis is carried out through “word cloud”. “R” programming is adopted to carry out these data analysis. Judgmental sampling is adopted in this study. They researched two movies “Bahubali” & “Udta Punjab”.
Jonas Krauss and Stefan Nann, identified three variables namely intensity, positivity and time that predicts the movie being nominated for the Academy Awards and thereby winning it. Though, winning an Oscar has always been speculative; the model was almost accurate in predicting if the movie would be nominated for the academy awards. The paper further demonstrates the approach by predicting the success of movies based on the communication in the online community IMDb.com. Authors also discuss the finding over three relevant components namely- Discussion Intensity, Positivity & Time.
RESULTS OF THE STUDY
- A. Factors that influence the frequency of watching a movie
- B. Understanding the probability of a movie success with Discriminant analysis
- C. Capturing audience sentiment using text mining analysis on Facebook, Movie Forums
The first component as discussed before tries to understand the different determinants of frequent movie watch. The data is obtained from a primary research involving variables such as Frequency of movie watch, online booking, Movie Purchase and ease of access. The model is developed by using linear regression analysis to identify the influencing factors for ‘Frequency’ of movie watch. “Frequency of movie going” is taken as dependent variable and the independent variables are “Online Booking”, “Movie Purchase”, “Ease of Access” and “peer preferences”. These independent variables are derived from the literature review and pilot test.
You can read the complete research article here.
Are you looking at building predictive models for your business, that help you plan procurement, operations to give you better control & ROI?