Answers are grounded in this paper and cite their sources.
Study showcases improved essay scoring under uncertainty with type 2 neutrosophic methods, indicating better assessment reliability.
Computing with uncertainty has become essential across numerous domains, including decision support systems, natural language understanding, and intelligent assessment, where incomplete, imprecise, or conflicting information often arises. In the field of education, one such challenge appears in automated essay scoring (AES), where linguistic ambiguity, subjective interpretation, and vague expressions introduce significant uncertainty. Traditional ontology-based AES systems attempt to structure domain knowledge to evaluate student essays, but they often fall short in handling these uncertainties effectively, resulting in reduced scoring reliability. To address this, we propose a novel approach that integrates Type-2 Neutrosophic Sets (T2NS) into ontology-driven AES frameworks. This integration enables a more refined representation of uncertain, indeterminate, and inconsistent information in essay evaluation. Our method involves concept extraction, rule-based scoring via ontologies, the extension of the ontology using T2NS, and the application of membership functions to quantify varying levels of uncertainty. Experiments conducted on benchmark AES datasets reveal that our T2NS-enhanced framework significantly outperforms conventional methods, offering more robust and interpretable scoring decisions under uncertainty.
No takes yet. Share an insight, caveat, or question.
Darwish et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: