Research and Application of Intelligent Analysis Technology for Vocational Education Job Requirements Based on Knowledge Graph and BiLSTM

Vocational education plays an increasingly important role in cultivating high-quality skilled talents. But how to accurately analyze and predict job demands has become an urgent problem to be solved in the current vocational education field. To optimize the existing talent demand matching analysis method, a job demand analysis method based on knowledge graph and BiLSTM is raised. The knowledge graph technology and BiLSTM neural network model were combined to build the professional career demand matching knowledge graph, and the crawler program was used to collect the college training program and industry job recruitment information in the network during the data acquisition stage of professional career demand matching knowledge graph. In the knowledge extraction stage, entity recognition of industry-related skill entities is carried out based on BiLSTM neural network and conditional random field entity recognition model. The experiment comparison outcomes denote that the entity query recall rate corresponding to the proposed method is higher than that of the other three job demand analysis methods, and is no less than 94%. The recall rates of the corresponding course name, skill name, job name and demand relationship are 96.3%, 95.2%, 96.6% and 96.4%, respectively. The experiment outcome indicate that the raised method has excellent performance in the comprehensive performance of entity recognition and entity query, and can play a better role in the analysis of job requirements.