Comparing the effectiveness of related articles recommender systems in Web of Science and Google Scholar

Abstract

Aim: Scientific recommender systems have been developed to provide users with resources that are close to their information needs. The main objective of the present study was to compare the effectiveness of the related articles recommender systems in Google Scholar and Web of Science databases from the users’ perspectives.Methods: This is an applied research with comparative approach. The samples included both human and paper ones. The human samples consisted of 120 Ph.D. candidates at Shiraz University. From among each field, thirty students (i.e., totally 120 students) participated in the study. The paper samples consisted of 2,400 related papers, 1200 of which were retrieved from Google Scholar, and 1,200 of which were retrieved from Web of Science database. The data were collected by a questionnaire and software.Results: The results showed that users considered both Google Scholar and Web of Science databases as effective in retrieving the related articles. There were significant differences between each of the four areas of humanities, basic sciences, engineering, agriculture and veterinary science in terms of the effectiveness of the related articles function. In addition, both databases represented the least number of related articles and the largest number of unrelated articles in the area of humanities. Originality: Despite the importance of related articles recommender systems, there was no evidence that measures the effectiveness of related articles recommender systems in Web of Science and Google Scholar.

Keywords

• user-oriented relevance
• effectiveness
• related articles recommender systems