Feature Selection in High Dimensional Biomedical Data Based on BF-SFLA.

Area of Science:

• Biomedical data analysis
• Bioinformatics
• Computational biology

Background:

• High-dimensional biomedical data presents challenges for disease diagnosis due to numerous irrelevant features.
• Feature selection is crucial for improving the efficiency and accuracy of diagnostic models.

Purpose of the Study:

• To develop and evaluate a novel feature selection method for high-dimensional biomedical data.
• To enhance disease diagnosis by optimizing feature subsets using an improved algorithm.

Main Methods:

• A novel feature selection method, the chemotaxis foraging-shuffled frog leaping algorithm (BF-SFLA), was proposed.
• The BF-SFLA incorporates chemokine operation and balanced grouping strategies to optimize global and local search capabilities.
• The method was evaluated using K-NN (k-nearest Neighbor) and C4.5 decision tree classification algorithms.

Main Results:

• The BF-SFLA demonstrated superior performance in selecting optimal feature subsets compared to existing methods.
• The proposed method significantly improved classification accuracy.
• Classification time was reduced using the BF-SFLA-based feature selection.

Conclusions:

• The BF-SFLA is an effective feature selection technique for high-dimensional biomedical data.
• This approach enhances diagnostic model performance by improving accuracy and reducing computational time.
• The method offers a promising solution for efficient disease diagnosis in bioinformatics.