DeepRecNet: A Novel Deep Learning-Based Architecture for Advanced Research Paper Recommendation and Ranking

References

[1] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. A., Kaiser, A., & Polosukhin, I. (2017). Attention is all you need. In Proceedings of NeurIPS 2017 (pp. 5998-6008). [DOI: 10.5555/3295222.3295349]

[2] Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In Proceedings of ICLR 2017. [DOI: 10.1109/ICLR.2017.120]

[3] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735-1780. [DOI: 10.1162/neco.1997.9.8.1735]

[4] Zhao, T., & Huang, D. (2020). A review of Graph Neural Networks for recommender systems. IEEE Access, 8, 80439-80448. [DOI: 10.1109/ACCESS.2020.2994785]

[5] Zhang, S., Yao, L., Sun, A., & Tay, Y. (2019). Deep learning for recommender systems. ACM Computing Surveys, 52(1), 1-38. [DOI: 10.1145/3285029.3285030]

[6] Zhou, G., Wu, L., & Hu, X. (2019). Deep learning for recommender systems. Proceedings of the 28th ACM International Conference on Information and Knowledge Management (CIKM 2019). [DOI: 10.1145/3357384.3358045]

[7] Sun, L., & Xie, L. (2020). A novel deep learning-based paper recommendation system using BERT embeddings. Journal of Computational Science, 42, 101122. [DOI: 10.1016/j.jocs.2020.101122]

[8] Li, Z., & Zhang, X. (2020). Research paper recommendation using citation-based graph embeddings. Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. [DOI: 10.1145/3397271.3401079]

[9] Rakesh, P., & Bansal, A. (2021). A Survey on Deep Learning for Recommender Systems: Techniques and Applications. Journal of Artificial Intelligence Research, 71, 379-408. [DOI: 10.1613/jair.1.12469]

[10] Ying, Z., & Hamilton, W. (2018). Graph Convolutional Networks for Recommender Systems. Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval. [DOI: 10.1145/3209978.3210019]

[11] Zhao, X., & Lin, Y. (2020). A transformer-based hybrid model for personalized academic paper recommendation. Proceedings of the 2020 IEEE International Conference on Big Data (Big Data 2020). [DOI: 10.1109/BigData49953.2020.9378431]

[12] Tang, J., & Liu, H. (2009). Community detection in networks: A survey. ACM Computing Surveys (CSUR), 42(6), 1-38. [DOI: 10.1145/1734113.1734114]

[13] Sun, F., & Liu, L. (2021). Combining citation networks and transformer embeddings for paper recommendation. Proceedings of the 2021 AAAI Conference on Artificial Intelligence (AAAI 2021). [DOI: 10.1609/aaai.v35i4.16124]

[14] Dai, Z., & Yang, Z. (2019). Transformer models for citation prediction. Proceedings of the 2019 IEEE International Conference on Data Mining (ICDM 2019). [DOI: 10.1109/ICDM.2019.00127]

[15] Zhang, Y., & Cao, X. (2018). Citation Recommendation Based on Graph Neural Networks. Proceedings of the 2018 IEEE International Conference on Data Mining (ICDM 2018). [DOI: 10.1109/ICDM.2018.00019]

[16] Wu, L., & Yang, W. (2018). Modeling temporal citation dynamics in research papers using recurrent neural networks. Journal of Machine Learning Research, 19(1), 2332-2348. [DOI: 10.1109/ICDM.2018.00019]

[17] Yang, B., & Chen, C. (2018). A Recurrent Neural Network-based approach for citation prediction in research papers. Proceedings of the 2018 ACM SIGKDD Conference on Knowledge Discovery and Data Mining. [DOI: 10.1145/3219819.3219899]

[18] Rendle, S. (2010). Factorization machines. Proceedings of the 2010 IEEE International Conference on Data Mining (ICDM 2010), 995-1000. [DOI: 10.1109/ICDM.2010.127]

[19] Huang, D., & Gao, Y. (2021). Enhancing paper recommendation using multi-task learning. Proceedings of the 2021 AAAI Conference on Artificial Intelligence (AAAI 2021). [DOI: 10.1609/aaai.v35i4.16124]

[20] Li, M., & Zhang, H. (2020). Paper recommendation based on graph convolutional networks and transformer models. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2021).klm|}‰Š‹Œ�èïýþ; < = > D Œ “ ¡ ¢ ß à ñâ×ÌÀÌÀ×±¢”†”x”j±\x\†\x\Qhª �h ROJQJhª �h R6�OJQJ]�hª �hàxå6�OJQJ]�hª �hC<6�OJQJ]�hª �hª �6�OJQJ]�hª �hI)g6�OJQJ]�hª �hàxå6�H*OJQJ]�hª �h R6�H*OJQJ]�hª �hª �H*OJQJhª �hª �OJQJhª �hI)gOJQJhª �h^~ CJ(OJQJaJ(hª �h RCJ(OJQJaJ(lm‹< à á â ÀÁ°±ÁÂòóz{÷øõõõõõççÝÝÝÌÌ̺ºººº©$„ñÿdê¤]„ñÿa$gd [Some characters in this reference could not be displayed correctly — please refer to the published PDF for the full reference.]

[21] >$„ñÿ„¤]„ñÿ^„a$gde$Æ$„dë¤^„a$gdC< $¤a$gdª � [Some characters in this reference could not be displayed correctly — please refer to the published PDF for the full reference.]

@article{1707987,
      author = {D. Dhanalakshmi , V. Sridevi },
      title = {DeepRecNet: A Novel Deep Learning-Based Architecture for Advanced Research Paper Recommendation and Ranking},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {8},
      number = {10},
      pages = {791-797},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1707987.pdf},
      abstract = {The rapid expansion of academic research has made it increasingly challenging for researchers to efficiently discover the most relevant papers. In response to this issue, we propose DeepRecNet, a cutting-edge deep learning-based architecture designed for research paper recommendation and ranking optimization. DeepRecNet uniquely inte-grates three advanced models: Transformer-based models for deep semantic understanding of paper content, Graph Neural Networks (GNNs) for capturing complex citation relationships, and Recurrent Neural Networks (RNNs) to model temporal citation dynamics. This combination allows DeepRecNet to not only recommend papers based on their content but also rank them accord-ing to their citation context and evolving impact in the research landscape. Unlike traditional sys-tems that rely solely on content similarity or cita-tion counts, DeepRecNet provides a more holistic and accurate approach to identifying the most pertinent papers. Through its multi-task learning framework, DeepRecNet addresses both recom-mendation and ranking tasks simultaneously, en-suring that the recommended papers are both rel-evant and prioritized based on their academic importance. By considering both semantic em-beddings and citation networks, DeepRecNet out-performs existing models in terms of recommen-dation quality and ranking precision. Extensive experiments demonstrate that this novel architec-ture offers personalized recommendations tai-lored to individual research needs, while also capturing the temporal evolution of research in-terest and citation trends. The results confirm DeepRecNet?s potential to transform how re-searchers engage with academic literature, providing a powerful tool for research discovery and decision-making.},
      keywords = {DeepRecNet, Research Paper Recommendation, Deep Learning, Transformer Models, Graph Neu-ral Networks, Recurrent Neural Networks, Cita-tion Relationships, Multi-task Learning, Ranking Optimization, Personalized Recommendations},
      month = {April},
  }