텍스트 마이닝을 이용한 신임경찰관 교육 개선방안 연구

Abstract

All institutions survey learners' satisfaction with their education after the end of educati on or training and use it as a resource to improve education. The Central Police Academy, which is in charge of training new police officers, also conducts a satisfaction survey at the end of on-campus training to improve the quality of education. However, the analysis of the educational satisfaction results is mainly centered on quantitative aspects such as a five-point scale, and the analysis of textual data containing various opinions of trainees is insufficient. In this study, we analyze the unstructured text data of the trainees' freely expressed opinions during the education satisfaction survey of the Central Police Academy for three years from 2019 to 2022 to suggest directions for improving the education of new police officers. For this purpose, 10 important words such as 'practice', 'field', and 'need' were identified based on the results of the analysis using TF-IDF after generating and preprocessing a custom dictionary that fits the characteristics of the police work domain. In addition, the similarity between words was measured using the Apori Algorithm and word2vec technique centered on the important words to check which words have high similarity with the important words to confirm the opinions of the trainees.