GA-Facilitated KNN Classifier Optimization with Varying Similarity Measures

Abstract- Genetic algorithms are powerful tools for k-nearest neighbors classifier optimization. While tradi-tional knn classification techniques typically employ Eu-clidian distance to assess pattern similarity, other mea-sures may also be utilized. Previous research demon-strates that GAs can improve predictive accuracy by searching for optimal feature weights and offsets for a cosine similarity-based knn classifier. GA-selected weights determine the classification relevance of each feature, while offsets provide alternative points of refer-ence when assessing angular similarity. Such optimized classifiers perform competitively with other contempo-rary classification techniques. This paper explores the effectiveness of GA weigh