EUVIMEDEUVIMED entdecken
Crossref

Hybrid Deep Learning Model for Evaluating Subjective Answers Based on Semantic Textual Similarity

Siddhesh Kudtarkar, Kavita Shirsat

Journal of Computer Science · 01.04.2026 · 22 (4) · 1189-1203

DOI: 10.3844/jcssp.2026.1189.1203 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computer Science, erschienen am 01.04.2026. Verfasst von Siddhesh Kudtarkar und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Subjective answer evaluation requires accurately identifying semantic similarities between student and reference responses. This study introduces a Hybrid Deep Learning Model (HDLM) that integrates CNN, GRU, and LSTM architectures to assess Semantic Textual Similarity (STS) more effectively. The HDLM employs two parallel branches-CNN-GRU and CNN-LSTM to capture both local syntactic features and long-range contextual dependencies, followed by Manhattan distance for semantic similarity computation. To address class imbalance and data sparsity, data augmentation and SMOTE resampling techniques are applied. The model is trained using the Quora Question Pairs dataset because of its substantial size and extensive semantic diversity. Comprehensive evaluation demonstrates HDLM’s superior performance (accuracy: 87.80%, F1-score: 0.88, AUC: 0.88) compared to existing models like Siamese LSTM, Multi-head Attention, and SOTA models such as SBERT and Sentence-T5. Statistical significance was validated using Wilcoxon signed-rank tests and 95% confidence intervals. Failure cases and case studies further highlight HDLM’s strengths and shortcomings. Overall, HDLM provides a robust, interpretable, and computationally efficient framework for automated subjective assessment.

Zitieren

APA

Kudtarkar, S., & Shirsat, K. (2026). Hybrid Deep Learning Model for Evaluating Subjective Answers Based on Semantic Textual Similarity. Journal of Computer Science, 22(4), 1189-1203. https://doi.org/10.3844/jcssp.2026.1189.1203

Vancouver

Kudtarkar S, Shirsat K. Hybrid Deep Learning Model for Evaluating Subjective Answers Based on Semantic Textual Similarity. Journal of Computer Science. 2026;22(4):1189-1203. doi:10.3844/jcssp.2026.1189.1203

BibTeX
@article{Kudtarkar2026,
  author = {Kudtarkar, Siddhesh and Shirsat, Kavita},
  title = {Hybrid Deep Learning Model for Evaluating Subjective Answers Based on Semantic Textual Similarity},
  journal = {Journal of Computer Science},
  year = {2026},
  volume = {22},
  number = {4},
  pages = {1189--1203},
  doi = {10.3844/jcssp.2026.1189.1203}
}
Kostenlos starten →