Some of the existing works have adapted highly advanced techniques instead of applying data mining methods to evaluate the box-office popularity. Electroencephalogram sensors were used to forecast the public responses for the movie based on the responses of cinema trailer views (Qing Wu∗, Wenbing Zhao∗, Himanshu Sharma∗, Tie Qiu). Due to limitations in the usage of the number of channels of EEG, the outcome was quite normal and there is a scope to incorporate machine learning models to attain accurate predictions. A movie recommender system was built by (RICHONG ZHANG) using Markov Chains factorization matrix process. Probabilistic matrix factorization and Singular-value decomposition (SVD++) were executed on the movie data collected to envisage the rating of the picture. The limitation of this model was implementing two MFMP models in basic forms and results could be good if constrained models were executed. The possible extension of the work could be considering manual data connected with picture ratings to grip cold-start and early-voter issues. Social network analysis and sentiment analysis has applied together to forecast the film popularity and how likely the movie has a chance of getting any awards (Krauss, Jonas; Nann, Stefan; Simon, Daniel). The main piece of this work was to analyse the public responses on the films from online and internet forums and also examining the historical awarded movies data. Then to examine is there any correlation between the online community reviews and film popularity at the box-office. The model has only considered few online forums and collecting a wide variety of sites may alter the results of the model. (Yong Liu) has analysed the word of mouth information and the data gathered from yahoo films online site to evaluate the popularity of the cinema at box-office. The sequels of the model determine that WOM actions are extremely effective during the pre-production stage and releasing week and the public will be highly interested in the opening week. So, this model only works on the attained reviews from the releasing weeks and failed in estimating the box office popularity during pre-release time. A predictive model was implemented by (Susmita S. Magdum) to forecast whether a film is a success or failure by mining the tweets from social media sites and public opinions from web bloggers. In this model, an Autoregressive Sentiment aware technique was used to roughly estimate the box office revenue and on classifying the positive and negative tweets a p-n ratio was considered to estimate movie success. And it also performed probabilistic latent semantic analysis to categorize the tweets and online comments. Due to lack of data, the model could not attain target results and by considering the huge amount of data and implementing Autoregressive sentiment and Quality Aware Model might produce accurate prediction in estimating movie popularity as well as profitability. Count based traits and content-based traits were collected from the online microblogs to predict the box-office popularity (Jingfei Du, Hua Xu). The relation between these features was analysed and on the cleansed data neural networks and SVM was executed to predict the popularity of the movie. The outcome could be improved by considering additional features because several factors are associated with the success of the film. Before the public release of a film, the cinema ongoing behaviour can be examined using Bass diffusion model and also taking words of mouth and opening week talks in to account (Feng WANG, Rong CAI, Minxue HUANG). The model produced a 60 per cent accuracy to estimate the film popularity in pre-production stage. The identified gap in this work is the lack of data which could be missing some additional factors that are influencing the success rate of picture and also achieved prediction rate could be improved by executing any conventional machine learning models. (Naghmeh Momeni1, Amir Tohidi Kalorazi2, Michael Rabbat1, and Babak Fotouhi) developed a model to forecast the movie profitability through an arbitrary model that associates micro-social synergies to macro observables in the case of dispersion of movie ongoing determinations. Through this epidemiological automatic model of film box-office dynamics, there is a probability of computing the likelihood of the viewer's decision and by examining those reviews the movie success will be categorized into certain ranges and can able to roughly estimate the gross revenue of the box-office.