Predicting IMDb Rating of TV Series with Deep Learning: The Case of Arrow

• Anna Luiza Gomes Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
• Getúlio Vianna Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
• Tatiana Escovedo Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio) http://orcid.org/0000-0002-7130-4330
• Marcos Kalinowski Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio) http://orcid.org/0000-0003-1445-3425

Resumo

Context: The number of TV series offered nowadays is very high. Due to its large amount, many series are canceled due to a lack of originality that generates a low audience. Problem: Having a decision support system that can show why some shows are a huge success or not would facilitate the choices of renewing or starting a show. Solution: We studied the case of the series Arrow broadcasted by CW Network and used descriptive and predictive modeling techniques to predict the IMDb rating. We assumed that the theme of the episode would affect its evaluation by users, so the dataset is composed only by the director of the episode, the number of reviews that episode got, the percentual of each theme extracted by the Latent Dirichlet Allocation (LDA) model of an episode, the number of viewers from Wikipedia and the rating from IMDb. The LDA model is a generative probabilistic model of a collection of documents made up of words. IS Theory: This study was developed under the aegis of Computational Learning Theory, which aims to understand the fundamental principles of learning and contribute to designing better-automated learning methods applied to the entertainment business. Method: In this prescriptive research, the case study method was used, and its results were analyzed using a quantitative approach. Summary of Results: With the features of each episode, the model that performed the best to predict the rating was Catboost due to a similar mean squared error of the KNN model but a better standard deviation during the test phase. It was possible to predict IMDb ratings with an acceptable root mean squared error of 0.55. Contributions and Impact in the IS area: The results show that deep learning techniques can be applied to support decisions in the entertainment field, allowing facilitating the decisions of renewing or starting a show. The rationale for building the model is detailed throughout the paper and can be replicated for other contexts.

Palavras-chave: Machine Learning, Deep Learning, LDA, prediction, tv-series, IMDb

Referências

Bristi, W.R., Zaman, Z. and Sultana, N., 2019, July. Predicting imdb rating of movies by machine learning techniques. In 2019 10th International Conference on Computing, Communication and Networking Technologies (ICCCNT) (pp. 1-5). IEEE.

Silge, J. and Robinson, D., 2017. Text mining with R: A tidy approach. O'Reilly Media, Inc.

Fitzgerald, T., 2020. How Many Is Too Many? There Are Now More Than 500 TV Shows. Available at [link]. Last access Jan 2020.

Sherman, A., 2020. Netflix has replaced broadcast TV as the center of American culture — just look at the viewership numbers. Available at [link]. Last access Jan 2020.

Alghamdi, R. and Alfalqi, K., 2015. A survey of topic modeling in text mining. Int. J. Adv. Comput. Sci. Appl.(IJACSA), 6(1).

Fronzetti Colladon, A. and Naldi, M., 2019. Predicting the performance of TV series through textual and network analysis: The case of Big Bang Theory. PloS one, 14(11), p.e0225306.

Akula, R., Wieselthier, Z., Martin, L. and Garibay, I., 2019, April. Forecasting the Success of Television Series using Machine Learning. In 2019 SoutheastCon (pp. 1-8). IEEE.

Blei, D.M., Ng, A.Y. and Jordan, M.I., 2003. Latent dirichlet allocation. the Journal of machine Learning research, 3, pp.993-1022.

Jelodar, H., Wang, Y., Yuan, C., Feng, X., Jiang, X., Li, Y. and Zhao, L., 2019. Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey. Multimedia Tools and Applications, 78(11), pp.15169-15211.

Cielen, D. and Meysman, A., 2016. Introducing data science: big data, machine learning, and more, using Python tools. Simon and Schuster.

EMC Education Services, 2015. Data science and big data analytics: discovering, analyzing, visualizing and presenting data. Wiley.

Maulud, D. and Abdulazeez, A.M., 2020. A Review on Linear Regression Comprehensive in Machine Learning. Journal of Applied Science and Technology Trends, 1(4), pp.140-147.

Alalousi, A., Razif, R., AbuAlhaj, M., Anbar, M. and Nizam, S., 2016. A preliminary performance evaluation of K-means, KNN and EM unsupervised machine learning methods for network flow classification. International Journal of Electrical and Computer Engineering, 6(2), p.778.

Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A.V. and Gulin, A., 2017. CatBoost: unbiased boosting with categorical features. arXiv preprint arXiv:1706.09516.

Zamani, M., Schwartz, H.A., Eichstaedt, J., Guntuku, S.C., Ganesan, A.V., Clouston, S. and Giorgi, S., 2020, November. Understanding weekly COVID-19 concerns through dynamic content-specific LDA topic modeling. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (Vol. 2020, p. 193). NIH Public Access.

Trstenjak, B., Mikac, S. and Donko, D., 2014. KNN with TF-IDF based framework for text categorization. Procedia Engineering, 69, pp.1356-1364.

Bengio, Y. and Grandvalet, Y., 2004. No unbiased estimator of the variance of k-fold cross-validation. Journal of machine learning research, 5(Sep), pp.1089-1105.

Escovedo, T. and Koshiyama, A., 2020. Introdução a Data Science: Algoritmos de Machine Learning e métodos de análise. Casa do Código.