International Journal of Computer Techniques Volume 12 Issue 5 | A Hybrid Neural Model for Identifying Similar Questions in Community Question Answering
A Hybrid Neural Model for Identifying Similar Questions in Community Question Answering
Author: Van-Tu Nguyen
Faculty of Natural Sciences and Technology, Tay Bac University, Son La, Vietnam
Email: tuspttb@utb.edu.vn
Journal: International Journal of Computer Techniques (IJCT)
Volume: 12 | Issue: 5 | Page: 71 | Publication Date: September – October 2025
ISSN: 2394-2231 | Journal URL: https://ijctjournal.org/
Abstract
This paper proposes a hybrid neural model for identifying semantically similar questions in community question answering (cQA) platforms. The model integrates Bidirectional Long Short-Term Memory (BLSTM) networks with a Multi-Layer Perceptron (MLP) and incorporates external knowledge features such as question type and category. Evaluations on SemEval 2016 Task 3 and Quora Question Pairs demonstrate competitive performance and consistent improvements over baseline models.
Keywords
Community Question Answering, Neural Network, Question Similarity, External Knowledge, BLSTM
Conclusion
The hybrid model effectively combines neural representations with external features to improve question similarity estimation. Future work will explore ablation studies and integration with advanced pre-trained language models for further enhancement.
References
Includes 27 references from SemEval, ACL, SIGIR, arXiv, and ACM covering neural architectures, semantic similarity, and question retrieval in cQA systems.
