User Profile in Personalized Service of Academic Digital Libraries: A Delphi Study

Objective: User profile is a key element in order to implement and develop personalization services. The user profile describes the user's preferences and lead to an understanding of the user's needs. The aim of this study was to identify indicators related to the creation and maintenance of user profiles and models in academic digital libraries.Methodology: The present study is an applied one done with the systematic review and Delphi method. Systematic review method was used to determine indicators related to creating and maintaining user profiles in the context of digital libraries. The primary keywords were searched in the different databases such as Google Scholar, Emerald, Ebsco, Scopus, Proquest, Magiran, Irandoc and Civilica. Search keywords include "Personali *", "Customiz *", "Recommendation System", "Personalized Recommendation" and combine them with the keywords "Librar *", "Digital Librar *", "Academic Librar *", "Services Librar *". In addition, the "My Library" keyword was searched. The scope of this systematic review included studies that were published from 1990 to 2019. Finally, 47 studies were selected on user profiles in personalized services of libraries. After reviewing the studies, the indicators on profile user were identified and then the Delphi method was used to determine important and basic indicators. Delphi's group is consisted of 15 experts in the field of digital library and library software who were selected using Purposeful sampling. Delphi's group reached a consensus on indicators after three rounds. Experts were asked to indicate the importance of the indicators using a 10-point scale ranging. The criterion score for the consensus of the experts was a mean of 7 or higher. The collected data were analyzed using SPSS version 16 software. Descriptive and inferential statistics including mean, standard deviation, binomial distribution and Kendall coefficient were used to analyze the data.Findings: 72 indicators were extracted to create and maintain user profiles. From the point of view of Delphi panel members, 49 indicators were recognized as important indicators. Important factors in creating and maintaining a user profile include the type of information, data collection approach, display and presentation of user modeling, user model interoperability, quality of data, privacy and security, and user profile management.Conclusion: Identifying indicators of user profiles can be used to develop and enhance the use of service pe