Selecting of the best features for the knn classification method by Harris Hawk algorithm
Publish place: 8th International Conference on New Strategies in Engineering, Information Science and Technology in the Next Century
Publish Year: 1400
نوع سند: مقاله کنفرانسی
زبان: English
View: 529
This Paper With 8 Page And PDF and WORD Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
EISTC08_032
تاریخ نمایه سازی: 15 اردیبهشت 1400
Abstract:
Today, due to the difficulty and complexity of selecting important and basic features from theinitial data, and also classifying the relevant features in terms of accuracy and training time,many methods in this field have been proposed by researchers that aim to All of them havebeen to simplify the selection of features and also increase the classification accuracy of therelevant features. Because performing the feature selection process improves classificationperformance and reduces training time and computational complexity. Therefore, in thispaper, the Harris Hawks Optimization (HHO) algorithm is described in order to extract themain features from the relevant data and remove duplicate features from the data, which fromthe k-nearest neighbor (knn) classification method is used as a fitness function to measure theaccuracy of classification.
Keywords:
Authors
Saeid Raziani
Department of Computer Engineering and Information Technology, Razi University Kermanshah, Iran
Taybeh Salehnia
Department of Computer Engineering and Information Technology, Razi University Kermanshah, Iran
Mahmood Ahmadi
Department of Computer Engineering and Information Technology, Razi University Kermanshah, Iran