THESIS
2016
xii, 124 pages : illustrations ; 30 cm
Abstract
With the rapid development of Web 2.0 and Online to Offline (O2O) marketing
model, various event-based social networks (EBSNs), such as Meetup, Plancast,
and Eventbrite, are getting popular. On EBSN platforms, event organizers organize
a variety of offline social events and online users register for and participate
in such offline events. An important task of EBSNs is to facilitate a satisfactory
event-participant arrangement for both sides, i.e. events enroll suitable participants
and participants are arranged with personally interesting events. Existing
approaches usually focus on arranging one single event to a set of potential
users and do not consider spatio-temporal information or conflicts among different
events, which can lead to infeasible arrangements. Also, no existi...[
Read more ]
With the rapid development of Web 2.0 and Online to Offline (O2O) marketing
model, various event-based social networks (EBSNs), such as Meetup, Plancast,
and Eventbrite, are getting popular. On EBSN platforms, event organizers organize
a variety of offline social events and online users register for and participate
in such offline events. An important task of EBSNs is to facilitate a satisfactory
event-participant arrangement for both sides, i.e. events enroll suitable participants
and participants are arranged with personally interesting events. Existing
approaches usually focus on arranging one single event to a set of potential
users and do not consider spatio-temporal information or conflicts among different
events, which can lead to infeasible arrangements. Also, no existing work considers
online scenarios of event arrangement, where users arrive at the platform one
by one and only partial information is available during the decision making process.
In addition, existing works measure the satisfaction of an arrangement by a
linear combination of few factors, weights of which are predefined and fixed, and
do not allow users to provide feedbacks on whether accepting the arrangement or
not. In this thesis, to address the shortcomings of existing approaches, we identify
more general and useful event-participant arrangement problems and propose efficient and effective solutions to address different scenarios of event-participant
arrangement over EBSNs. To summarize, our study addresses the following four
problems:
• We identify an event-participant arrangement problem, called the Global
Event-participant Arrangement with Conflict and Capacity (GEACC) problem,
that focuses on resolving conflicts of different events and making event-participant
arrangements in a global view. We prove that the GEACC
problem is NP-hard and design quality-guaranteed approximate algorithms
to address this problem.
• We further address the online scenario of GEACC and design an online
algorithm with provable performance guarantee.
• We study another event-participant arrangement problem that further considers
location information of events and users. We present a greedy-based
heuristic algorithm and a two-step approximation framework with guaranteed
approximation ratio and a series of optimization techniques.
• We study another online scenario of event-participant arrangement where
users who arrive in an online way can choose to accept or reject the arranged
events and the satisfaction of an arrangement is learned adaptively.
We model the problem as a contextual combinatorial bandit problem and
present efficient and effective algorithms to solve the problem.
We verify the effectiveness and efficiency of the proposed solutions with extensive
experiments and discuss interesting future work on event-participant arrangement over EBSNs.
Post a Comment