Application of Rough Ant Colony Algorithm in Adolescent Psychology

Area of Science:

• Computer Science
• Criminology
• Artificial Intelligence

Background:

• Big data research is crucial in the security protection industry.
• Predicting juvenile delinquency is a significant challenge.
• Understanding the psychological factors in juvenile crime is essential.

Purpose of the Study:

• To propose a novel algorithm for predicting juvenile delinquency.
• To address limitations of existing classification algorithms.
• To leverage big data for enhanced security and crime prevention.

Main Methods:

• Developed a Rough Ant Colony (RoughAC) classification algorithm.
• Utilized rough set theory for membership degree determination (upper and lower approximate sets).
• Integrated membership values into the ant colony algorithm's pheromone update mechanism.

Main Results:

• RoughAC effectively solves premature convergence and stagnation near local optima.
• The algorithm handles continuous domain and combinatorial optimization problems.
• Achieved superior classification results in predicting juvenile delinquency tendencies.

Conclusions:

• The RoughAC algorithm offers a robust approach to juvenile delinquency prediction.
• This method provides valuable insights for crime prevention and intervention.
• The integration of rough sets and ant colony optimization enhances classification accuracy.