以人工智慧技術建構職業訓練課程推介模型研究

Study on AI Technology Application for the Establishment of the National Vocational Training Program

本研究係以勞動部所屬機關之公務資料為基礎，規劃以大數據分析及人工智慧建置職業訓練課程推介模型，提供有效資訊並同時最佳化訓練課程效益。基於研、訓、檢、用的核心理念，改善訓及用之間的落差，引進人工慧技術分析過去職業訓練課程參與者特質及其訓後就業情形，並進一步根據職訓課程參與者特質及其訓後就業情形進行分組分類，使用演算法進行計算，建立課程推介模型。本研究將勞保資料及職業訓練資訊系統資料進行串聯、整理過後以機器學習演算法根據勞工是否參與職業訓練進行分類分群，找出影響勞工參與職業訓練的特徵值，再根據定義之就業成效指標計算參訓勞工之訓後就業成效，根據相關結果以人工智慧建置職業訓練課程推介模型。職業訓練課程推介模型以資料為基礎，目的在於建構可針對不同勞工特性推介適合的職業訓練課程模型。本研究主要應用決策樹演算法建立職業訓練課程受訓者分類模型及職業訓練課程受訓者就業成效預測模型。

Based on the data owned by the agencies of the Ministry of Labor, this research plans to use the big data analysis and artificial intelligence to build a vocational training curriculum promotion model to provide effective information and optimize the effectiveness of the training curriculum. Based on the core concepts of research, training, inspection and use, improve the gap between training and use, introduce artificial intelligence technology to analyze the characteristics of past vocational training course participants and their post-training employment situation, and further establish a classification model based on the vocational training curriculum. The traits and their post-training employment situation are grouped into categories, and then the algorithm is used for calculation to establish a curriculum recommendation model. In this study, we serialized and organized the labor insurance data and the vocational training information system data, and then used the machine learning algorithm to classify and group according to whether the workers participated in vocational training, and found out the characteristic values that affect the labor participation in vocational training. The defined employment effectiveness indicators are used to calculate the post-training employment effectiveness of the participating laborers. Based on the relevant results, the vocational training curriculum promotion model is built with artificial intelligence. The vocational training curriculum promotion model is based on information and aims to create a model that can be used to recommend suitable vocational training courses for different workers based on their characteristics. In this study, the decision tree algorithm is applied to establish the classification model of the trainees of the vocational training courses and the prediction model of the employment performance of the trainees of the vocational training courses.