Abstract
A lot of daily activities require more than one person to participate and collaborate with each other; however, for many people, it is not easy to find good partners to engage in activities with one another. With the rapid growth of social network applications, more and more people get used to creating connections with people on the social network. Therefore, designing social network framework for partner-matching is significant in helping people to easily find good partners. In this paper, we proposed a framework which can match partners for an active community. In order to improve the matching performance, all users are divided into groups based on a specific classification tree that is built for a specific activity. The optimization goal of the partner-matching is to maintain as many stable partnerships as possible in the community. To achieve the goal, various factors are considered to design matching functions. The simulation results show that the proposed framework can help most people find stable partners quickly.
Highlights
A lot of daily activities require two or more people to collaborate
A lot of people liking these activities suffer from finding good partners
We proposed a partner-matching framework to address this issue
Summary
A lot of daily activities require two or more people to collaborate. For instance, playing tennis, squash, rock climbing, and ballroom dancing. Our proposed framework designs to use a classification tree to distribute users to leaf nodes of the tree; for a user’s request, the searching pool is way smaller compared to finding the best match among all candidates. Algorithms of stable roommate problem cannot be applied to partner-matching applications since giving a preference list will violate the rule of protecting privacy of users.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
