Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{170458,
        author = {Kandhati Saivarsha and Mandala Saikumar and Kati Caleb and Konda Raveendra kumar},
        title = {Can we predict student performance based on tabular and textual data},
        journal = {International Journal of Innovative Research in Technology},
        year = {2024},
        volume = {11},
        number = {7},
        pages = {333-335},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=170458},
        abstract = {With the rise of teaching systems like MOOCs, massive amounts of educational data, including student behavior metrics and course comments, are being generated but remain underutilized for uncovering useful models for school management. To address this, we collected a multi model data set combining tabular student behavior data with textual course comments and proposed a Transformer-based framework to fuse these data types into a uniform vector representation for predicting student performance. Empirical results show that our approach significantly improves prediction accuracy, with F1-scores increasing by up to 3.33% and AUC by up to 4.37% compared to existing methods. Validation on an open data set confirmed the framework’s strong generalization capability. Using the SHAP method for interpret-ability, we found that textual features have a greater influence on the classification model, further demonstrating the value of integrating text data. These findings suggest that fusing behavioral and textual features not only improves model performance but also provides actionable insights for educational data mining.},
        keywords = {},
        month = {December},
        }