Mingon Kang (Computer Science) published a paper titled, "Hi-LASSO: High-performance python and apache spark packages for feature selection with high-dimensional data," in PLOS ONE in collaboration with Gyeongsang National University in South Korea. The graduate student authors at Gyeongsang National University conducted this research under the supervision of Kang at UNLV between 2019 and 2020. The paper introduces Python and Apache Spark libraries of Hi-LASSO, which is a bootstrapping-based LASSO model for high-dimensional data feature selection. The libraries enable test of significance on the feature selection with high-dimensional but low sample size data.
Share your thoughts about this story. To comment, you'll need to login into your Facebook account. Your comment will post immediately. Comments that are not in keeping with our comment policies may be removed by editors.