Attendance Maximization for Successful Social Event Planning

TL;DR

This paper addresses the challenge of maximizing social event attendance by proposing scalable algorithms for the Social Event Scheduling problem, which is proven to be NP-hard, and demonstrates significant efficiency improvements over existing methods.

Contribution

The paper introduces three novel algorithms for the NP-hard Social Event Scheduling problem, improving computational efficiency and scalability in maximizing event attendance.

Findings

Algorithms perform on average half the computations of existing solutions.

Proposed algorithms are 3-5 times faster in several cases.

Extensive experiments validate the effectiveness and scalability of the algorithms.

Abstract

Social event planning has received a great deal of attention in recent years where various entities, such as event planners and marketing companies, organizations, venues, or users in Event-based Social Networks, organize numerous social events (e.g., festivals, conferences, promotion parties). Recent studies show that "attendance" is the most common metric used to capture the success of social events, since the number of attendees has great impact on the event's expected gains (e.g., revenue, artist/brand publicity). In this work, we study the Social Event Scheduling (SES) problem which aims at identifying and assigning social events to appropriate time slots, so that the number of events attendees is maximized. We show that, even in highly restricted instances, the SES problem is NP-hard to be approximated over a factor. To solve the SES problem, we design three efficient and scalable algorithms. These algorithms exploit several novel schemes that we design. We conduct extensive experiments using several real and synthetic datasets, and demonstrate that the proposed algorithms perform on average half the computations compared to the existing solution and, in several cases, are 3-5 times faster.