HTTPS SSH

Keystroke feature extractor

Extract fixed-length feature vectors from a raw keystrokes file. The keystrokes file should be in csv format with columns:

  user, session, timepress, timerelease, keyname

The script creates a feature file with columns:

  user, session, <feature1>, <feature2>, ...

Usage

python main.py extract [-h] [--features FEATURES] [--measure MEASURE]
                       [--fallback FALLBACK] [--normalization NORMALIZATION]
                       [--norm_distance NORM_DISTANCE] [--groupby GROUPBY]
                       input output

positional arguments:
  input                 Raw keystroke data file
  output                Feature file output

optional arguments:
  -h, --help            show this help message and exit
  --features FEATURES   Feature set
  --measure MEASURE     Feature measure function
  --fallback FALLBACK   Fallback hierarchy
  --normalization NORMALIZATION
                        Normalization method
  --norm_distance NORM_DISTANCE
                        Std distance for std normalivation
  --groupby GROUPBY     Groupby columns