Personality Prediction for State Grid E-Learning Text Based on Neural Networks with Attention Mechanism

State Grid Corporation of China (SGCC), as the leading enterprise in China’s power industry, is well aware that employees are the core strength of enterprise development, and therefore has always attached great importance to employee training. For employee training of SGCC, informatization education can break limitations of time and locations, allowing employees to learn anytime and anywhere and meeting different learning needs. State Grid E-Learning, as an online learning platform of SGCC, plays a crucial role in informatization education and training of SGCC. The platform has accumulated amounts of online user data. among which the unstructured text accounts for a large portion, and the used text often shows the natural state and true expression in different scenarios. Understanding the personality traits of employees helps improve their satisfaction, loyalty, and performance. Therefore, in this study, for the Big Five Personality Traits of Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism, two models with different internal principles are constructed based on State Grid E-Learning text using variants of two classic neural network structures, respectively. The Dilated Gated Convolutional Neural Network with Attention (A-DGCNN) model is constructed based on the convolutional neural network with dilated gated convolution and attention mechanism; and the Bidirectional Long Short-Term Memory with Attention (A-Bi-LSTM) model is constructed based on the bidirectional long short term memory of the recurrent neural network and attention mechanism. This work provides reference value for predicting other related personality traits and tendencies.