Automatic Scoring of Students’ Science Writing Using Hybrid Neural Network

Ehsan Latif, Xiaoming Zhai

Track: Innovations in AI for Education (Day 1)

Paper Length: long-paper (6 pages + references)

Keywords: Science Education, Hybrid Neural Network (HNN), Automatic Scoring, Machine Learning, and Educational Technology

TL;DR: Its an analysis paper to evaluate the performance of Hybrid Neural Network for Multi-perspective automatic scoring of science education student responses

Abstract: This study explores the efficacy of a multi-perspective hybrid neural network (HNN) for scoring student responses in science education with an analytic rubric. We compared the accuracy of the HNN model with four ML approaches (BERT, ANN, Naive Bayes, and Logistic Regression). The results have shown that HHN achieved 8%, 3%, 1%, and 0.12% higher accuracy than Naive Bayes, Logistic Regression, ANN, and BERT, respectively, for five scoring aspects (p < 0.001). The overall HNN’s perceived accuracy (M = 96.23%, SD = 1.45%) is comparable to the (training and inference) expensive BERT model’s accuracy (M = 96.12%, SD = 1.52%). We also have observed that HNN is x2 more efficient in terms of training and inferencing than BERT and has comparable efficiency to the lightweight but less accurate Naive Bayes model. Our study confirmed the accuracy and efficiency of using HNN to automatically score students’ science writing.

Submission Number: 58