This website implements a reviewer-article assignment method based on topic models. In this way, the assignment only depends on the article's abstracts and sample abstracts from potential reviewer. This different from other methods that require either authors suggesting reviewers or reviewers bidding for articles to review.
Once the reviewer's score for each article are in, we have an estimator of the paper's score. This estimator controls for the discrepancies in variance and mean score of reviewers --- reviewers' scores that are systematically too harsh or too soft will be moved to the mean.
If you have any questions, please do not hesitate to ask Titipat Achakulvisut or other members of the team.