Profiling high leverage points for detecting anomalous users in telecom data networks
- PDF / 2,136,071 Bytes
- 9 Pages / 595.224 x 790.955 pts Page_size
- 30 Downloads / 181 Views
Profiling high leverage points for detecting anomalous users in telecom data networks Shazia Tabassum1
˜ Gama3 · Muhammad Ajmal Azad2 · Joao
Received: 22 November 2019 / Accepted: 15 July 2020 © Institut Mines-Tlcom and Springer Nature Switzerland AG 2020
Abstract Fraud in telephony incurs huge revenue losses and causes a menace to both the service providers and legitimate users. This problem is growing alongside augmenting technologies. Yet, the works in this area are hindered by the availability of data and confidentiality of approaches. In this work, we deal with the problem of detecting different types of unsolicited users from spammers to fraudsters in a massive phone call network. Most of the malicious users in telecommunications have some of the characteristics in common. These characteristics can be defined by a set of features whose values are uncommon for normal users. We made use of graph-based metrics to detect profiles that are significantly far from the common user profiles in a real data log with millions of users. To achieve this, we looked for the high leverage points in the 99.99th percentile, which identified a substantial number of users as extreme anomalous points. Furthermore, clustering these points helped distinguish malicious users efficiently and minimized the problem space significantly. Convincingly, the learned profiles of these detected users coincided with fraudulent behaviors. Keywords Fraud detection · Unsolicited users · Anomaly detection
1 Introduction Profiling is a technique to identify behavioral patterns of users based on some properties available in a specific context. They are also referred to as signatures or patterns in the literature [1, 2]. These profiles are usually constructed using data available from the past or current (direct or indirect) interactions with the system. The users with similar behavior can share a common profile with a smaller variance relative to dissimilar behaviors. User profiling is exercised in numerous applications of various domains for customization and improvement of services, decision making, recommendation systems, etc. One of the fields where it was primarily applied was in telecommunications [3], where some of the applications include advertising products, promotions, segmentation of packages and catalogs, Shazia Tabassum
[email protected] 1
INESC TEC, University of Porto, R. Dr. Roberto Frias, Porto, Portugal
2
University of Derby, Derby, UK
3
INESC TEC, Porto, Portugal
and development of service infrastructure, etc. Besides, user profiles have been persistently used in detecting fraudulent or unsolicited users [4]. Fraudsters realize the widespread use of telecommunications as a potential platform to target victims. The initiation of fraud over Internet technologies, specifically over the telephone networks, is cheaper and easier than committing fraud in an offline society [5]. In the telephone networks, fraudsters can target both the service providers as well as end-users. Frauds targeting service providers could
Data Loading...
