A Distributed Graph Algorithm for Discovering Unique Behavioral Groups from Large-Scale Telco Data

Page view(s)
16
Checked on Feb 25, 2024
A Distributed Graph Algorithm for Discovering Unique Behavioral Groups from Large-Scale Telco Data
Title:
A Distributed Graph Algorithm for Discovering Unique Behavioral Groups from Large-Scale Telco Data
Journal Title:
CIKM '16 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management
Keywords:
Publication Date:
24 October 2016
Citation:
Abstract:
It is critical for a large telecommunications company such as Singtel to truly understand the behavior and preference of its customers, in order to win their loyalty in a highly fragmented and competitive market. In this paper we propose a novel graph edge-clustering algorithm (DGEC) that can discover unique behavioral groups, from rich usage data sets (such as CDRs and beyond). A behavioral group is a set of nodes that share similar edge properties reflecting customer behavior, but are not necessarily connected to each other and therefore different from the usual notion of graph communities. DGEC is an optimization-based model that uses the stochastic proximal gradient method, implemented as a distributed algorithm that scales to tens of millions of nodes and edges. The performance of DGEC is satisfactory for deployment, with an execution time of 2.4 hours over a graph of 5 million nodes and 27 million edges in a 8-machine environment (32 cores and 64GB memory per machine). We evaluate the behavioral groups discovered by DGEC by combining other information such as demographics and customer profiles, and demonstrate that these behavioral groups are objective, consistent and insightful. DGEC has now been deployed in production, and also shows promising potential to extract new usage behavioral features from other data sources such as web browsing, app usage and TV consumption.
License type:
PublisherCopyrights
Funding Info:
Description:
Full paper can be downloaded from the Publisher's URL provided.
ISBN:
978-1-4503-4073-1
Files uploaded:
File Size Format Action
There are no attached files.