Elderly Behavior Analysis to Organize Appropriate Rehabilitation Activity Packages Using Data Mining Techniques

Abstract

The purposes of this research article were 1) to study and collect data on rehabilitation activities for the elderly in Mueang District, Chiang Mai Province; and 2) to organize appropriate rehabilitation activities for the elderly using the K-Mode Clustering technique. This was the survey research. The sample consisted of 400 elderly individuals from 4 communities in Mueang District, Chiang Mai Province. This survey study was conducted using a stratified random sampling technique. Data were collected using a content validity questionnaire with a reliability of 0.90. Data were analyzed using descriptive statistics, Chi-square, and clustering analysis using K-Mode Clustering technique. The research results found that 1) the elderly needed a variety of activities according to their basic factors and health status. They were divided into 5 activity groups based on the concept of “5 Happiness”: health, recreation, integrity, cognition, and peacefulness; 2) K-Mode Clustering technique divided the elderly into 4 subgroups in each activity group. The Health Group had an average Silhouette Score of 0.182 and consisted of 25.8% of the elderly, with 68.4% liking exercise activities. The Recreation Group had an average Silhouette Score of 0.187 and consisted of 19.4% of the elderly, with 71.2% enjoying music and art activities. The Integrity Group had an average Silhouette Score of 0.172 and consisted of 16.1% of the elderly, with 64.3% liking social service activities. The Cognition Group had an average Silhouette Score of 0.184 and consisted of 22.6% of the elderly, with 75.5% liking learning activities. And the Peacefulness Group had an average Silhouette Score of 0.175 and had 16.1 percent of the elderly, with 69.8 percent liking mental relaxation activities.