Research on the Design of Question-label Classification Model Based on LSTM Network

With the promotion of the whole process and business of the digital transformation of State Grid, the requirements for the digital ability of each business department are getting higher and higher, and higher training and evaluation requirements are also put forward on how to quickly familiarize the company’s personnel with the business and master the operation and production skills. In order to conduct quantitative evaluation and analysis of the skill mastery of the trainees and recommend precise training courses, this paper established a question-label classification model based on the LSTM(Long Short Term Memory) network. This model was based on the training question-label system constructed by manual labeling method, and it combined with the training evaluation work of the State Grid of China Technology College. It realizes the automatic classification and annotation of the imported test questions.