A study of tag-based recipe recommendations for users in different age groups

Abstract

Social tagging becomes prevailing with the emergence of Web 2.0 communities recently. By utilizing this additional valuable information from user-created tags, it is convenient to understand users’ interests and behavior so that we can provide good user experience in applications of various domains. Therefore, the definition of profile is crucial for tagging systems. Furthermore, it is important to have recommendations for various groups of users. In recipe recommendations, older people typically have different needs compared with young users. In this paper, we first focus on the definitions of user profile, item feature and how to derive semantics from these sources. Afterwards, we design the framework of the tag-based multimedia recipe recommendation system (MRRS). Finally, we conduct preliminary experimental study of recipe recommendations for different age groups. The result shows that older people concern more about the nutrition aspects of recipes.