Implementing a Sparse Logistic Regression Algorithm in Apache Spark