People-to-people networks play very important roles in our daily lives. This paper presents an in-depth study of such networks and develops an efficient match-making system. To this end, the authors integrate tools and techniques from online dating networks, social network analysis (SNA), graph theory, and match-making systems, which involve psychological factors and recommendations based on user ratings, similarities in user interests, and relationships with other users.

As a technical tool, the authors use attributed bipartite graphs where vertices represent users with associated attributes such as age, gender, interests, occupation, and education; edges represent the relationships between users. The authors then employ various SNA techniques and discuss corresponding implications. The discussion is informative and nicely written for easy reading. Using network data represented by an attributed bipartite graph, the proposed model identifies communities in order to make recommendations.

Empirical analysis shows that the proposed system outperforms other systems on most known benchmarks and can generate recommendations in a reasonable amount of time. I recommend this paper to readers interested in graph theory and network design.