Spark classification models

Spark classification models.

Table of content

System and experiment settings

Summary of results

Linear classification models

Two classification learning methods will be discussed, support vector machines SVM and logistic regression LR. The application context is single label binary classification. They can also be applied to single label multiclass classification which however will not be covered in this blog post.

Load and save data files

Support vector machine SVM (code)

Run SVM with parameter selections

Model test

Experimental results

Logistic regression LR (code)

Run LR for parameter selection

Experimental results

External reading materials

Hongyu Su 18 October 2015