Decision Support Systems

Decision Support Systems (DSS)
Name
Course
Instructor
Date
Business intelligence systems case study
The case study focuses on forecasting the success of films and movies by utilizing data mining and analytics. There is a lot of uncertainty on predicting the financial performance of the movie industry, and hence Sharda and Delen propose a model that explores the problem. Sharda and Delen (2006) utilize neural networks to test revenue estimates through box office receipts by classifying films into nine categories including ‘flop’ to blockbuster’ to improve the results of the model, the authors do not rely on a single neural network but rather use cross validation to obtain more robust results. The results of the experiment show that the ensemble model should be relied upon than individual models (Turban et al., 2013).
Hollywood decision makers and data mining
Hollywood decision makers should consider data mining as there is need for more quality data sources to show better pattern analysis. It is through data mining that the decision makers can use predictors to build models that forecast box-office receipts (Olson & Delen, 2008). The movie industry is unpredictable and improving certainty through utilizing data mining improves the industry’s financial success through evaluating trade offs in movie production. Additionally, the predictive models are important, as industry players can invest in movie production, and emphasize on movies that are likely to be successful, while reducing budgetary allocations to movies that have impressive results and this improves the industry’s returns on investment (ROI).
Challenges facing Hollywood managers and similar industry segments
The top challenge for Hollywood managers is get the best results after investing scarce resources in movie production. The managers need to get the best ROI and assess likely success of movie production, but with limited prediction capabilities this becomes hard to achieve. The managers invest in movies to get returns, but choosing the optima mix of investments remains a challenge. Predicting the demand of movies hinders the managers’ ability to maximize returns of risky investments in the form of movie production. The managers also rely on instincts where there is lack of enough data to support their claims, and at times these instincts are proven wrong. The sports industry that relies on ticketing also faces similar challenges, and the food and fashion industries have difficult to predict demand.
Relevance of prediction models and classification problem
The researchers used data mining to evaluate patterns and then focused on classification and regression. They relied on all the relevant data available with box office grosses revenue being the dependent variable. This is the most important factor in the prediction, and by leaving out auxiliary revenues, t...