| 英文摘要 |
This study aims to apply Educational Data Mining (EDM) techniques to explore the potential characteristics and structural patterns of in-service training courses for high school teachers serving as administrators, utilizing large-scale training data. The research data were obtained from the Ministry of Education’s ''National In-service Teacher Education Information Network,'' comprising completed training records for the year 2024. The dataset includes 5,442 high school teachers serving as administrators, with a total of 104,390 course participation entries. The study employs the Apriori algorithm to mine association rules between teacher backgrounds and course attributes. By utilizing Support, Confidence, and Lift, meaningful and positively correlated rules were filtered to serve as a reference for improving the planning of in-service training courses for teachers. The results of the study indicate the following: 1. Demographic Dominance: The primary group of high school teachers serving as administrators participating in in-service training consists of teachers from public schools and those aged 40 to 50. 2. Course Preferences: These teachers show a strong preference for courses categorized as ''Level 1: Basic–Understanding and Familiarity'' and short-duration courses of ''less than 2 hours.'' 3. Regional Variations: Significant regional differences in training characteristics were observed: the Northern region tends toward short-duration courses; the Central region shows a prevalence of training in public schools; the Southern region exhibits distinct gender-based participation patterns; and the Eastern region prefers daytime sessions. |