<strong>Who is Dating Whom: User Behavior Characterization, Prediction, and Recommendation for Online Dating</strong>

<strong>Who is Dating Whom: User Behavior Characterization, Prediction, and Recommendation for Online Dating</strong>

时间：2015年12月17日上午9：30 地点：九龙湖计算机楼313会议室

报告简介：

Online dating sites have become popular platforms for people to look for potential romantic partners, offering an unprecedented level of access to potential dates that is not available through traditional means. We study how users’ online dating behaviors correlate with various user attributes using a large real-world dateset from a major online dating site in China. Many of our results align with notions in social and evolutionary psychology: males tend to look for younger females while females place more emphasis on the socioeconomic status (e.g., income, education level) of a male. We observe that the geographic distance between two users and the photo count of users play an important role in their dating behavior, and it is important to differentiate between users’ true preferences and random selection. We also find that both males and females are more likely to reply to users whose attributes come closest to the stated preferences of the receivers, but there is significant discrepancy between a user’s stated dating preference and his/her actual online dating behavior. For online dating, it is important that not only the recommended profiles should match the user’s preference, but also the recommended users should be interested in the user and thus likely to reciprocate when contacted. We study the reply prediction problem and approach it using a machine learning framework. Our results show that user-based and graph-based features result in similar performance, and can be used to effectively predict the reciprocal links. We further propose a reciprocal recommendation system, in which we measure the mutual interest and attractiveness between a pair of users with different similarity functions and make recommendations based on the similarity scores. The results show that our proposed algorithms significantly outperform previous approaches.

报告人简介：

Dr. Benyuan Liu is an Associate Professor in the Department of Computer Science at the University of Massachusetts Lowell. He received his BS degree in physics from University of Science and Technology of China (USTC), MS degree in physics from Yale University, and PhD degree in computer science from University of Massachusetts Amherst. Dr. Liu's main research interests are in the area of application, algorithm design and performance analysis of various computer networking technologies. His research results have been published in premium computer networking conferences and journals. His research has been supported by the National Science Foundation (NSF), National Institutes of Health (NIH), DARPA, and Microsoft Research. He is a recipient of the NSF CAREER Award.

计算机网络和信息集成教育部重点实验室（东南大学）