@inproceedings { , title = {Towards explainable metaheuristics: PCA for trajectory mining in evolutionary algorithms.}, abstract = {The generation of explanations regarding decisions made by population-based meta-heuristics is often a difficult task due to the nature of the mechanisms employed by these approaches. With the increase in use of these methods for optimisation in industries that require end-user confirmation, the need for explanations has also grown. We present a novel approach to the extraction of features capable of supporting an explanation through the use of trajectory mining - extracting key features from the populations of NDAs. We apply Principal Components Analysis techniques to identify new methods of population diversity tracking post-runtime after projection into a lower dimensional space. These methods are applied to a set of benchmark problems solved by a Genetic Algorithm and a Univariate Estimation of Distribution Algorithm. We show that the new sub-space derived metrics can capture key learning steps in the algorithm run and how solution variable patterns that explain the fitness function may be captured in the principal component coefficients.}, conference = {41st British Computer Society's Specialist Group on Artificial Intelligence (SGAI) Artificial intelligence international conference 2021 (AI-2021)}, doi = {10.1007/978-3-030-91100-3\_7}, isbn = {9783030910990}, note = {INFO COMPLETE (Now published, checked and updated 21/12/2021 LM; Info of acceptance from contact 16/9/2021 LM) PERMISSION GRANTED (version = AAM; embargo = 12 (from publication); licence = Pub's own; POLICY = https://resource-cms.springernature.com/springer-cms/rest/v1/content/15433008/data/Contract\_Book\_Contributor\_Consent\_to\_Publish\_LNCS\_SIP ) DOCUMENT READY (AAM rec'd from contact 16/9/2021 LM) ADDITIONAL INFO - Contact: MARTIN FYVIE; John McCall; Lee Christie Set Statement - (The final authenticated version is available online at: https://doi.org/10.1007/978-3-030-91100-3\_7. This pre-copyedited version is made available under the Springer terms of reuse for AAMs: https://www.springer.com/gp/open-access/publication-policies/aam-terms-of-use )}, pages = {89-102}, publicationstatus = {Published}, publisher = {Springer}, url = {https://rgu-repository.worktribe.com/output/1457054}, volume = {13101}, keyword = {Computational Intelligence (CI), Evolutionary algorithms, PCA, Explainability, Population diversity}, year = {2021}, author = {Fyvie, Martin and McCall, John A.W. and Christie, Lee A.} editor = {Bramer, Max and Ellis, Richard} }