Feature Selection with Harmony Search

Ren Diao, Qiang Shen

Research output: Contribution to journalArticlepeer-review

141 Citations (SciVal)


Many search strategies have been exploited for the task of feature selection (FS), in an effort to identify more compact and better quality subsets. Such work typically involves the use of greedy hill climbing (HC), or nature-inspired heuristics, in order to discover the optimal solution without going through exhaustive search. In this paper, a novel FS approach based on harmony search (HS) is presented. It is a general approach that can be used in conjunction with many subset evaluation techniques. The simplicity of HS is exploited to reduce the overall complexity of the search process. The proposed approach is able to escape from local solutions and identify multiple solutions owing to the stochastic nature of HS. Additional parameter control schemes are introduced to reduce the effort and impact of parameter configuration. These can be further combined with the iterative refinement strategy, tailored to enforce the discovery of quality subsets. The resulting approach is compared with those that rely on HC, genetic algorithms, and particle swarm optimization, accompanied by in-depth studies of the suggested improvements.
Original languageEnglish
Pages (from-to)1509-1523
Number of pages15
JournalIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
Issue number6
Publication statusPublished - Dec 2012


Dive into the research topics of 'Feature Selection with Harmony Search'. Together they form a unique fingerprint.

Cite this