基于COMET的现代学徒制职业能力聚类分析

Cluster Analysis of Occupational Competency in Modern Apprenticeship Based on COMET

针对区分、筛选、分析职业能力的定性研究方法存在提取的典型工作任务与企业具体工作过程匹配度不高,以及对现代学徒制针对性不强等问题,提出了一种定量的职业能力聚类分析方法。采用COMET能力模型区分出能力内容、工作过程、能力要求三种职业能力属性,识别出职业能力的频率、重要性和关联性三种特征,构建了职业能力的六维属性数据库;提出基于DBSCAN算法的聚类有效性评估方法,确定最优聚类数后采用两步算法进行聚类分析;提取每一簇的特征词,并计算其六维属性的分布、总值和均值,最终归纳提炼出典型工作任务。分析结果明确了职业能力的属性及其内在联系,更加真实的反映了企业的实际工作情景与要求。

In distinguishing,screening and analyzing occupational competency,qualitative research method has some drawbacks such as low match between typical work task and the specific work process in the enterprise,and low relevancy to the modern apprenticeship system.A quantitative clustering analysis method is proposed.The COMET competency model is used to extract three types of occupational competency:competency components,work process and competency requirement.Then three characteristics including frequency,importance and relevancy of the competency are identified to construct six dimensional database of occupational competency.A clustering validity evaluation method based on DBSCAN algorithm is proposed.After the optimal number of clusters,cluster analysis was carried out by using two-step algorithm.Then typical words of each cluster are extracted,followed by calculation of the distribution,gross value and mean of these words’six dimensional properties.Typical work tasks are finally extracted.The results of the analysis clearly define the attributes and internal relations of occupational competency,which more objectively reflects the actual work situation and requirements of the enterprise.