Learning to Assess the Quality of Scientific Conferences: A Case Study in Computer Science

Assessing the quality of scientific conferences is an important and useful service that can be provided by digital libraries and similar systems. This is specially true for fields such as Computer Science and Electric Engineering, where con-ference publications are crucial. However, the majority of the existing approaches for assessing the quality of publica-tion venues has been proposed for journals. In this paper, we characterize a large number of features that can be used as criteria to assess the quality of scientific conferences and study how these several features can be automatically com-bined by means of machine learning techniques to effectively perform this task. Within the features studied are citations, submission and acceptance rates, tradition of the confer-ence, and reputation of the program committee members. Among our several findings, we can cite that: (1) separat-ing high quality conferences from medium and low quality ones can be performed quite effectively, but separating the last two types is a much harder task; and (2) citation fea-tures followed by those associated with the tradition of the conference are the most important ones for the task