The Evaluating Word and Sentence Embeddings for Automatic Short Answer Grading

Authors

Abstract

Recent advances in Artificial Intelligence (AI) and Deep Network (DN) enable natural language processing (NLP) has empowered Automated short Answer Grading (ASAG) models. Products are now more versatile across domain independent applications. The user however still sees them as operational black boxes and where specific fine tuning is not possible. Results, evaluated and compared over several standard metrics brings out the performance measures in clearly fathomable units. The word empowers technologically uninitiated users, mostly in evaluation to make an educated choice of the technical best suited for ASAG of a specific context based on pre-existing indicators like language ability and native language.

Downloads

Download data is not yet available.

Published

2026-08-01

How to Cite

Das, S. (2026). The Evaluating Word and Sentence Embeddings for Automatic Short Answer Grading. International Journal of Linguistics, Literature and Culture, 12(5). Retrieved from https://sloap.org/journals/index.php/ijllc/article/view/2615

Issue

Section

Research Articles