University Links: Home Page | Site Map
Covenant University Repository


Ogunleye, O. and Ifebanjo, T. and Abiodun, T. and Adebiyi, A. A. (2017) PROPOSED FRAMEWORK FOR A PAPER-REVIEWER ASSIGNMENT SYSTEM USING WORD2VEC. In: CUCEN 2017, Covenant University, Ota.

[img] PDF
Download (960kB)


The task of assigning papers to reviewers is crucial to the realization of a peer-to-peer review process of academic conferences. The manual process of ensuring submissions assigned to reviewers is related to their knowledge domain can be very cumbersome. Besides, poor quality reviews results from an ineffective assignment of papers. From extant literature, automated reviewer assignment systems based on distributional semantic models have been used to capture semantics with the shortcoming of limited in the bag of words models. Neural Network Language models have been used to eliminate the limitations of bag of words of models in expertise finding and product recommendation. Thus this paper proposes a framework based on neural network language models to derive suitability scores based on the semantic relatedness between a paper meant for review and a reviewer’s representation papers. The present performance of the neural network language model compared to distributional semantic models used in solving reviewer-assignment. This ensures the semantic relatedness of paper and reviewer knowledge representation in allocating a paper, which improves the overall success of the peer-to-peer review process.

Item Type: Conference or Workshop Item (Paper)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science
Depositing User: Mr Adewole Adewumi
Date Deposited: 05 Mar 2018 19:14
Last Modified: 05 Mar 2018 19:14

Actions (login required)

View Item View Item