Free cookie consent management tool by TermsFeed Policy Generator

Opened 6 years ago

Last modified 6 years ago

#2990 accepted feature request

Variable-Impact-based Feature Selection

Reported by: pfleck Owned by: pfleck
Priority: medium Milestone: HeuristicLab 3.3.17
Component: Algorithms.DataAnalysis Version: branch
Keywords: Cc:

Description

Using the variable impacts for iterative feature selection can currently only be done manually. Having an automatic way would be nice.

General workflow of this Variable-Impact-based Feature Selection:

  1. Train Model
  2. Calculate Variable Impacts
  3. Remove Features with low impact
  4. Repeat until model accuracy drops too much

Change History (3)

comment:1 Changed 6 years ago by pfleck

  • Status changed from new to accepted
  • Version set to branch

r16612 branched trunk

comment:2 Changed 6 years ago by pfleck

r16705: Implemented first version with RF and percentage-based feature elimination.

comment:3 Changed 6 years ago by abeham

  • Milestone changed from HeuristicLab 3.3.16 to HeuristicLab 3.3.17
Note: See TracTickets for help on using tickets.