Design of an Intelligent Network English Automatic Scoring Algorithm based on CS-ELM
Jie Qin · 2024
In order to solve the problem of traditional English scoring relying on manual evaluation, this paper proposed an intelligent network English automatic scoring algorithm called CS-ELM (Compressive Sensing-Extreme Learning Machine) to overcome the challenges of traditional methods being time-consuming and easily influenced by subjective factors. The article provided a detailed introduction to the principle and characteristics of the CS-ELM algorithm, which is a single hidden layer extreme learning machine based on competition mechanism, with fast training and good generalization ability. The CS-ELM was applied to English automatic scoring tasks, transforming English writing samples into inputs for the CS-ELM model through feature extraction and selection. After experimental testing, the root mean square error of the algorithm model in this article was between 0.01 and 0.06. CS-ELM shows good performance in English automatic scoring tasks, with lower root mean square error and high correlation coefficient.