A Comprehensive Survey of Time Series Anomaly Detection in Online Social Network Data

Abstract

In the field of data mining, the social network is one of the complex systems that poses significant challenges in this area. Time series anomaly detection is one of the critical applications. Recent developments in the quantitative analysis of social networks, based largely on graph theory, have been successfully used in various types of time series data. In this paper, we review the studies on graph theory to investigate and analyze time series social networks data including different efficient and scalable experimental modalities. We provide some applications, challenging issues and existing methods for time series anomaly detection.

Discovering dynamic adverse behavior of policyholders in the life insurance industry

“…Existing research has pointed out that traditional techniques are rather time-consuming, taking up to several months, and it is costly to capture comprehensive information on policyholder behaviors. Therefore, it is important for IMs to remain alert for changes, demands, and necessary actions to manage and work with local industries (Bolhaar et al, 2012;Islam et al, 2017;Keane and Stavrunova, 2016). However, the major challenge for IMs is to keep track of the behavioral patterns of policyholders.…”