Skip to content

Feature Extraction by Grammatical Evolution for One-Class Time Series Classification

License

Notifications You must be signed in to change notification settings

spaghettix/FE_GE_TSC

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

53 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Feature Extraction by Grammatical Evolution for One-Class Time Series Classification

Run feature_extraction.py to extract features from your time series data-set.

The file run_a_feature_extractor.py shows how to transform your time series in a feature-based representation according to a given feature-extractor. Then, the file shows how to evaluate classification performance in a one-class scenario.

Other important files:

  1. Gramamr -> /grammars/tsc_grammar.bnf
  2. Evolutionary parameters -> /parameters/tsc_parameters.txt
  3. Fitness function -> /src/fitness/tsc_fitness.py
  4. Functions used in the grammar (primitives) -> /src/fitness/math_functions.py

Please refer to the paper for further information. Get in touch with me if necessary.

Contact: mauceri.stefano@gmail.com

Note: This repository is basically a copy of PonyGE2 [2] with some adjustments required to perform feature-extraction from time series as described in the related paper [1].

References:

[1] Mauceri, Stefano, James Sweeney, Miguel Nicolau, and James McDermott. "Feature extraction by grammatical evolution for one-class time series classification." Genetic Programming and Evolvable Machines (2021): 1-29.

[2] Fenton, Michael, James McDermott, David Fagan, Stefan Forstenlechner, Erik Hemberg, and Michael O'Neill. "Ponyge2: Grammatical evolution in python." In Proceedings of the Genetic and Evolutionary Computation Conference Companion, pp. 1194-1201. 2017.

Releases

No releases published

Packages

No packages published

Languages