An efficient ontology-based topic-specific article recommendation model for best-fit reviewers
- 30 Downloads
In general peer review is accredited as the vital and utmost cornerstone of the scientific publishing and research developments. Undeniably, the reviewers play a decisive role in ensuring the qualitative scientific developments published in any venue (Journals, conferences). The conventional time-tested method of double-blind peer review has been criticized having the flaws of inability to find the novelty, paucity of clarity, paucity of soundness, prone to be biased, the paucity of impartiality, discrepancies amongst reviewers, the paucity of acknowledgment and inspiration to reviewers. In order to cope with some of its flaws and to ensure the excellence of peer review, it is indispensable to delve into the process of article recommendation to the best fit reviewers. Typically, this recommendation is done by the human expert, so less accurate as the manual recommendation is incapable of initial scrutinizing of the tome of articles submitted and best fitting reviewer’s profile. This work proposes ontology and topic-specific personalized recommendation system to recommend the articles to the best-fit reviewers. In this proposed ontology-based model, latent semantic analysis and entropy have been deployed for similarity measure and topic-specificity indicator, thus to fetch the information of the best-fitted reviewer’s profile. In this work, an experimental arrangement has been set up relying on the primary dataset related to the reviewer’s profile and article reviewed. Results show the feasibility of the proposed model and the correlational relationship between the semantics and the topic-specificity of the articles which could be adopted as an automatic article recommendation to best fitting reviewers.
KeywordsAutomatic personalized recommendation (APR) LSA Entropy measure Correlational analysis Ontology Bibliometric
This work was supported by National Key R&D Program of China 2018YFD1100300.
- Abbasi, R., Rehman, G., Lee, J., Riaz, F. M., & Luo, B. (2017). Discovering temporal user interest on Twitter by using the semantic-based dynamic interest finding model (TUT). In 2017 IEEE 2nd information technology, networking, electronic and automation control conference (ITNEC) (pp. 743–747). https://doi.org/10.1109/ITNEC.2017.8284832.
- Chughtai, G. R., Lee, J., Kabir, A., et al. (2018a). Ontology-based academic article recommendation. In ICCWAMTIP (pp. 93–96). Chengdu: IEEE Xplore. https://doi.org/10.1109/ICCWAMTIP.2018.8632601.
- Rehman, G., & Lee, J. (2017). Quantifying the impact of hot-paper on new researchers. In ICCIS 2017 proceedings of the 2017 2nd international conference on communication and information systems (pp. 329–334). Wuhan, China—November 07–09, 2017. ACM. https://doi.org/10.1145/3158233.3159323.
- Shah, N., & Song, Y. (2015). s-index: Towards better metrics for quantifying research impact. Retrieved from arXiv:1507.03650.