Downloads: 109 | Views: 176
Research Paper | Computer Science | United States of America | Volume 9 Issue 6, June 2020
Sparse Bayesian Machine Learning with application to NBA Data (Python & R)
Yu Wang [2] | Yaoxuan Luan [2]
Abstract: Machine Learning is one of the hot searches in the search engine and is very useful in a variety of areas and subjects. The definition of Machine learning given by Wikipedia is the study of computer algorithms that improve automatically through experience. Mathematical equations and statistical computations also play crucial roles in the entire machine learning process. Statistical models like regression and classification help in Supervised Learning. Other processes will work for Unsupervised Learning and Reinforcement Learning. Other than pure statistics, which focus more on understanding data in terms of models, Machine Learning focus higher on prediction. This article focusses on the prediction perspective of the Machine Learning process while considering the dimension reduction using the sparsity property. The LASSO (Tibshirani, 1996) method provides a sharp power in selecting significant explanatory variables and has become very popular in solving big data problems. A simulation study was conducted to test the power of the model. For application, NBA data was considered. A prediction of the 2019 postseason bracket is given by learning the historical postseason team statistics. The accuracy of the bracketing could be evaluated.
Keywords: Machine Learning, Bayesian, Sparsity, Dimension Reduction
Edition: Volume 9 Issue 6, June 2020,
Pages: 577 - 580