A Novel Approach on Conducting Reviewer Recommendations Based on Conflict of Interest

the reviewer's recommendation in accordance with the field of research is crucial where this is directly proportional to the results of the review on the research. In finding suitable reviewers, conflicts of interest (COI) are often found. In this paper, we propose an approach to reviewer recommendations with topic extraction and author extraction to prevent COI. First, we separate the process into 2, namely to do topic extraction using Latent Dirichlet Allocation (LDA), and also to do author extraction using Cosine Similarity. The next step is to combine the two results to rank the 10 recommended authors. The experimental results show that our approach has succeeded in getting 10 recommended reviewers to avoid COI.