From: Predicting academic success in higher education: literature review and best practices
Methods | When to use | Advantages | Disadvantages |
---|---|---|---|
Vertical selection | To remove redundant or irrelevant features | Facilitate understanding of the extracted pattern and rises the speed of the learning stage | Requires a good understanding of the data domain |
Horizontal selection | To remove redundant and/or conflicting instances | Enhance the quality of input data, thus enable improved performance of DM models | In descriptive analysis, it is best to have as many instances as possible. |