An Association Rule-Based Recommendation Engine for an Online Dating Site

Abstract

Being a popular social network type, online dating sites provide a platform for people to find partners for establishing a relationship. In this study, a recommendation engine for one of the prominent online dating sites of Turkey is developed. It works as a support system to suggest potential matches to the site users. As opposed to the traditional systems that match users based on their revealed preferences, the engine is based on a rule set extracted from the past communication data, using association rule mining. A list of best matches based on scoring derived from these rules is presented. The performance of the engine is statistically tested. It is found that the scores of matching couples are found to be significantly higher than the non-matched couples' scores.