A NOVEL AND HYBRID WHALE OPTIMIZATION WITH RESTRICTED CROSSOVER AND MUTATION BASED FEATURE SELECTION METHOD FOR...

Area of Science:

• Computational intelligence
• Machine learning for healthcare
• Swarm intelligence applications

Background:

• Anxiety and depression are prevalent psychological disorders with significant health impacts.
• Accurate diagnosis is crucial for effective treatment and management.
• Current diagnostic methods can benefit from advanced computational approaches.

Purpose of the Study:

• To develop and evaluate an optimal feature selection technique for diagnosing anxiety and depression.
• To compare the efficacy of swarm intelligence-based metaheuristic algorithms for psychological disorder diagnosis.
• To introduce a novel hybrid algorithm, Restricted Crossover Mutation based Whale Optimization Algorithm (RCM-WOA).

Main Methods:

• A dataset of 1128 instances and 46 attributes related to depression and anxiety was utilized.
• Nine metaheuristic techniques including Genetic Algorithm, Grey Wolf Optimizer, and Whale Optimization Algorithm were employed.
• A novel hybrid feature selection method, RCM-WOA, was designed to balance exploration and exploitation.

Main Results:

• Swarm intelligence algorithms were applied and evaluated using metrics like accuracy, sensitivity, and specificity.
• The proposed RCM-WOA achieved a diagnostic accuracy of 91.4%.
• Performance metrics demonstrated the effectiveness of RCM-WOA in identifying optimal feature sets.

Conclusions:

• Anxiety and depression are critical disorders requiring effective diagnostic tools.
• The RCM-WOA demonstrates superior performance compared to existing state-of-the-art methods for diagnosing these conditions.
• This research offers a promising computational approach for improving the diagnosis of psychological disorders.