Designing an expert system for fraud detection in private telecommunications networks

Hilas, Constantinos S.

doi:10.1016/j.eswa.2009.03.031

Cited by 59 publications

(25 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They have been used in the telecommunication industry to support engineers in the design of local area networks [35,36], rural radio access networks [37], cable networks [38], network management applications [39][40][41], and optical fiber communication links [34]. However, those for 60 GHz RoF land network design appear to be non-existent to our best knowledge.…”

Section: Introductionmentioning

confidence: 96%

Knowledge-based framework for the design of millimeter-wave (60 GHz) radio over fiber land networks

et al. 2015

View full text Add to dashboard Cite

This paper presents the research work involved in the development of a knowledge-based framework for the design of millimeter-wave 60 GHz radio over fiber (RoF) land networks. It combines object-oriented, rule-based, technical information, and procedural functions to support engineers in the conceptual and preliminary design of a network. The overall framework is organized in two main modules: the first module is a tutorial of the terminology and basic principles in the design of RoF networks. The second module is a design assistant system that requests input data from the user about the functional network requirements and its prioritized figures of merit. The proposed design assistant is capable of providing support on the major activities of the 60 GHz RoF land network design such as downlink and uplink channel assignment and network clustering, wavelength allocation, optical link design, and network integration. The assistant is user interactive through the implementation of a graphical user interface. To illustrate the validation of such framework, we present a case of study of a network design with specific requirements.

show abstract

Section: Introductionmentioning

confidence: 96%

Knowledge-based framework for the design of millimeter-wave (60 GHz) radio over fiber land networks

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Hilas and Sahalos [10] presented a similar approach in several ways. Their main concentration is to construct certain profiles using PBX-based calls, that are used to determine whether a user is suspicious of becoming fraudulent.…”

Section: Previous Workmentioning

confidence: 99%

Non-conforming behavior detection for VoIP-based network systems

Galiotos

Anagnostopoulos

Dagiuklas

et al. 2016

2016 IEEE International Conference on Communications (ICC)

View full text Add to dashboard Cite

Abstract-This work proposes a detection scheme that identifies non-conforming behavior in a VoIP network, based on statistical analysis and hypothesis testing. VoIP networks are a popular, low-cost alternative for telephony that offer lower rates especially for long-distance calls. Other services such as FollowMe, enhance the traditional voice-oriented nature of these networks. Consequently several security concerns such as fraud calls, are related to the high availability required by a VoIP system. Fraud calls account for an average loss of 3% to 5% of the operators' revenue. Thus the detection and prevention of the users from behaving in a non-conforming way, becomes crucial. A trustworthy and secure management and billing scheme is necessary, to guarantee the proper operation. This work proposes a behavioral control scheme for the VoIP clients. An initial training period defines the normal behavior. Then statistical analysis and t-testing is employed to extract results regarding the users' profiles, with pre-defined confidence levels. A Buffer zone creates a more flexible decision-making process. The scheme also offers the ability to configure its parameters, in order to react appropriately under different network conditions and detect possible misuses. It is implemented and its rational operation is verified via several simulation scenarios.

show abstract

“…Fraud detection in the telco context aims at detecting individuals that acquire a mobile phone and do not intend to pay their contract [8]. Typical approaches focus on classifying users according to their level of risk by calculating deviations from standard behaviors [20,11]. Telco user models have also been effectively used for the improvement of the network infrastructure, including the design of pricing plans, an application where mobility data has proven extremely relevant.…”

Section: Related Workmentioning

confidence: 99%

“…The number of calls and total talk time can each be further restricted according to direction of call, where each subscriber's incoming and outgoing calls are considered separately. From an application perspective, variations of these features are very relevant, for example the ratio between national and international calls or between calls made within the provider and outside the provider are very relevant for churn [9,2] and fraud detection [20,11]. Also these two variables are relevant to viral marketing as users that have a lot of connections are more capable of spreading information [10].…”

Section: Features Of Telco User Modelsmentioning

confidence: 99%

User Modeling for Telecommunication Applications: Experiences and Practical Implications

Hohwald

Frı́as-Martı́nez

Oliver

2010

User Modeling, Adaptation, and Personalization

View full text Add to dashboard Cite

Abstract. Telecommunication applications based on user modeling focus on extracting customer behavior and preferences from the information implicitly included in Call Detail Record (CDR) datasets. Even though there are many different application areas (fraud detection, viral and targeted marketing, churn prediction, etc.) they all share a common data source (CDRs) and a common set of features for modeling the user. In this paper we present our experience with different applications areas in generating user models from massive real datasets of both mobile phone and landline subscriber activity. We present the analysis of a dataset containing the traces of 50, 000 mobile phone users and 50, 000 landline users from the same geographical area for a period of six months and compare the different behaviors when using landlines and mobile phones and the implications that such differences have for each application. Our results indicate that user models for a variety of applications can be generated efficiently and in a homogeneous way using an architecture based on distributed computing and that there are numerous differences between mobile phone and landline users that have relevant practical implications.

show abstract

Designing an expert system for fraud detection in private telecommunications networks

Cited by 59 publications

References 13 publications

Knowledge-based framework for the design of millimeter-wave (60 GHz) radio over fiber land networks

Knowledge-based framework for the design of millimeter-wave (60 GHz) radio over fiber land networks

Non-conforming behavior detection for VoIP-based network systems

User Modeling for Telecommunication Applications: Experiences and Practical Implications

Contact Info

Product

Resources

About