<![CDATA[The selection of extracurricular activities is an important element in developing students' character, interests, and potential outside of academic activities. However, this selection process is often not carried out systematically, so it risks not being in accordance with individual student preferences. This study aims to design an extracurricular recommendation model based on the Particle Swarm Optimization (PSO) algorithm, one of the intelligent computing methods that is effective in finding optimal solutions. The model considers four main criteria, namely interests, talents, schedules, and student personalities, which are formulated into an objective function. The PSO algorithm is used to find the best combination of extracurricular activities for each individual based on predetermined preference weights. Testing using data from 50 students shows that the PSO algorithm is able to produce recommendations with the best fitness value that is stable at around 4.5 after the 40th iteration, indicating the algorithm's effectiveness in finding optimal solutions. Correlation analysis between variables shows a strong positive relationship between Interests and Talents (correlation coefficient approaching 0.75), while Personality and Schedule contribute unique information with a correlation below 0.2. The results of this study demonstrate that the PSO algorithm approach is proven to be an adaptive and efficient approach that can be used as an effective method to generate personalized and relevant extracurricular recommendations. It can also be used as a tool in decision-making in educational environments and increase student participation and potential development in a more targeted and optimal manner. The test results show that this model is able to provide more personalized and appropriate recommendations compared to conventional methods. Furthermore, the PSO algorithm provides a relatively fast convergence time with high decision accuracy.]]>
Copyrights © 2025
