Testing the Fraud Detection Ability of Different User Profiles by Means of FF-NN Classifiers (original) (raw)
Related papers
User Profiling for Fraud Detection in Telecommunication Networks
Telecommunications fraud is increasing dramatically each year resulting in loss of a large amount of euros worldwide. A statistical machine learning method is presented that constructs user profiles for the detection of fraudulent activities in telecommunications networks. The approach presented here can be used for the detection of superimposed or hacking fraud, works well for mid-term decisions and cannot be used for on-line account comparison.
Comparing the Performance of Some Neural Fraud Detectors in Telecommunications
Fraud detection in telecommunications is an area where pattern recognition and so called "intelligent" techniques have found widespread use. Due to fraud, companies suffer not only direct economic losses but also the risk of bad publicity. In this paper real cases of fraud are being treated in order to develop a detection system with low number of false alarms and good sensitivity. Call data records provide a number of measures that can be used to discriminate fraudulent activities from correct ones. Three neural networks schemes have been applied to such data comparing latter their results with new cases.
Australian Journal of Basic and Applied Sciences , 2010
Neural Networks is an essential information-processing paradigm,that is inspired by a way of biological nervous systems, which can be used in predicting fraud occurrence in mobile phone usage. Superimposed fraud occurrence could occur as a result of overlapping calls and irregularities in the time pattern spent in calling. The power of neural network-based technology is a potent mechanism in combating the menace of superimposed fraud in mobile technology. The methodology employed in this research work included data collection by survey from a telecommunication industry in Africa, data testing and analysis by making use of a Neural Network Software known as NeuroSolutions. Performance comparative analysis was carried out by using six different neural network models. The final deductions from the results of the experiments carried out, showed that the Fuzzy network model outperformed the other five neural networks in terms of the least error difference generated between the predicted output and the final output generated. This showed that Fuzzy network model is more efficient in its performance in comparison with the other five models.
Predicting Fraud in Mobile Phone Usage Using Artificial Neural Networks
Journal of Applied Sciences Research, 2008
Mobile phone usage involves the use of wireless communication devices that can be carried anywhere, as they require no physical connection to any external wires to work. However, mobile technology is not without its own problems. Fraud is prevalent in both fixed and mobile networks of all technologies. Frauds have plagued the telecommunication industries, financial institutions and other organizations for a long time. The aim of this research work and research publication is to apply 3 different neural network models (Fuzzy, Radial Basis and the Feedforward) to the prediction of fraud in real-life data of phone usage and also analyze and evaluate their performances with respect to their predicting capability. From the analysis and model predictability experiment carried out in this scientific research work, it was discovered that the fuzzy network model had the minimum error generated in its fraud predicting capability. Thus, its performance in terms of the error generated in this fraud prediction experiment showed that its NMSE (Normalized mean squared error) for the fraud predicted was 1.98264609. The mean absolute error (MAE = 15.00987244) for its fraud prediction was also the least; this showed that the fuzzy model fraud predictability was much better than the other two models.
Advanced Machine Learning Technologies and Applications, 2021
Cybercrimes and fraud techniques are major threats to telecommunications sectors in the last decade, one of those fraud approaches called Wangiri fraud. Wangiri is a common type of fraud techniques in telecommunications sector, the definition is originated from a Japanese word that means (one & cut); as the fraudsters depend on a single ring method to gain illegal money from the subscribers. Consequently, the approaches that are used to detect fraud cases are used to classify subscribers based on their behaviors such as data extraction that identifies patterns in large datasets through a combination of statistical methods, artificial intelligence and databases. Neural networks are used to process & evaluate the given datasets; in order to uncover ambiguous communication and secret data patterns. This paper proposes the usage of AI neural networks to overcome the highly predictive wangiri fraud in a telecom dataset and to make an effective and convenient classification. ISTAT was used to test both accuracy and efficiency of the proposed method.
Telecommunications Subscription Fraud Detection using Artificial Neural Networks.pdf
Telecommunications Companies are facing a lot of problems due to fraud; hence the need for an effective fraud detection system for the telecommunications companies. This paper presents a design and implements of a subscription fraud detection system using Artificial Neural Networks. Neurosolutions for Excel was used to implement the Artificial Neural Network. The system was tested and found to be user friendly, effective and 85.7% success rate achieved.
Telecommunications Subscription Fraud Detection using Artificial Neural Networks
Telecommunications Subscription fraud detection using Artificial Neural Networks , 2015
Telecommunications Companies are facing a lot of problems due to fraud; hence the need for an effective fraud detection system for the telecommunications companies. This paper presents a design and implements of a subscription fraud detection system using Artificial Neural Networks. Neurosolutions for Excel was used to implement the Artificial Neural Network. The system was tested and found to be user friendly, effective and 85.7% success rate achieved.
Subscription fraud prevention in telecommunications using fuzzy rules and neural networks
Expert Systems with Applications, 2006
A system to prevent subscription fraud in fixed telecommunications with high impact on long-distance carriers is proposed. The system consists of a classification module and a prediction module. The classification module classifies subscribers according to their previous historical behavior into four different categories: subscription fraudulent, otherwise fraudulent, insolvent and normal. The prediction module allows us to identify potential fraudulent customers at the time of subscription. The classification module was implemented using fuzzy rules. It was applied to a database containing information of over 10,000 real subscribers of a major telecom company in Chile. In this database a subscription fraud prevalence of 2.2% was found. The prediction module was implemented as a multilayer perceptron neural network. It was able to identify 56.2% of the true fraudsters, screening only 3.5% of all the subscribers in the test set. This study shows the feasibility of significantly preventing subscription fraud in telecommunications by analyzing the application information and the customer antecedents at the time of application.
Knowledge Based Systems, 2008
This paper investigates the usefulness of applying different learning approaches to a problem of telecommunications fraud detection. Five different user models are compared by means of both supervised and unsupervised learning techniques, namely the multilayer perceptron and the hierarchical agglomerative clustering. One aim of the study is to identify the user model that best identifies fraud cases. The second task is to explore different views of the same problem and see what can be learned form the application of each different technique. The models are compared in terms of their performances. Each technique's outcome is evaluated with appropriate measures.
Fraud Detection using Neural Network
Recent studies indicate that in certain people with the risk of carrying cash; plastic money has become the trend these days. Past history of transactions like card holders' spending pattern from the previous transaction database, and other inputs like income, location, living expense etc can be compared with the current transaction details to detect credit card frauds. Deviation on transactions can indicate a fraud. There have been a few studies conducted to determine how to detect frauds by using different technology.