Abstract
Current information recommendation systems obtain users’ preferences from Web browsing histories and activities such as purchase of products, and efficiently provide the users with their preferable information. In such a case, however, the same or similar information is always recommended, which is called filter bubble and it decreases the users’ satisfaction to the systems. If information recommendation systems could provide users with something surprising and useful as output information, the user’s satisfaction to the systems would drastically increase. Therefore, “serendipity” is paid attention to in this research. In this paper, a new information recommendation system using a concept-based information retrieval is proposed to provide the users with serendipitous information. In this system, concepts which describe features or roles of items are input instead of the items themselves, and information which can meet the concepts are output as candidates of serendipitous information. The serendipitous information is extracted from the output information using the criteria which are the indexes of serendipity defined in this research. Through the evaluation experiment, it is revealed that the proposed system achieves the accuracy of 70% for the serendipitous information determination and the accuracy of 100% for the information retrieval, which are satisfactory for this research purpose.
Highlights
In the current information society, recommendation systems are widely utilized to efficiently provide the users with preferable information extracted from a huge amount of resources
In this paper, an approach to obtain serendipitous information when users perform information retrieval was discussed based on the proposed indexes of the serendipity, which are "relevance", "unexpectedness" and "usefulness"
The system achieved the serendipitous information determination accuracy of 70%
Summary
In the current information society, recommendation systems are widely utilized to efficiently provide the users with preferable information extracted from a huge amount of resources. Useful information for the users are recommended. Pariser (2011) and Jauhar (2015) stated that recommendation systems can recommend favorite information for the users, those information are unnecessary in most cases. This is because such recommended information is always similar to the ones recommended before. If only preferable information is recommended to the users, it causes a filter bubble problem in which information in other categories cannot be recommended
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Turkish Journal of Computer and Mathematics Education (TURCOMAT)
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
