Telecom fraud detection with big data analytics

Massive Computing, 2002

Finding telecommunications fraud in masses of call records is more difficult than finding a needle in a haystack. In the haystack problem, there is only one needle that does not look like hay, the pieces of hay all look similar, and neither the needle nor the hay changes much over time. Fraudulent calls may be rare like needles in haystacks, but they are much more challenging to find. Callers behavior for another, while all needles look the same. Moreover, fraud has to be found repeatedly, as fast as fraud calls are placed, the nature of fraud changes over time, the extent of fraud is unknown in advance, and fraud may be spread over more than one type of service. For example, calls placed on a stolen wireless telephone may be charged to a stolen credit card. Finding fraud is like finding a needle in a haystack only in the sense of sifting through masses of data to find something rare. This paper describes some issues involved in creating tools for building fraud systems that are accurate, able to adapt to changing legitimate and fraudulent behavior, and easy to use.

International Journal of Engineering Research and Technology (IJERT), 2012

https://www.ijert.org/1-a-model-for-rule-based-fraud-detection-in-telecommunications https://www.ijert.org/research/1-a-model-for-rule-based-fraud-detection-in-telecommunications-IJERTV1IS5337.pdf Telecommunications fraud is a worldwide problem that deprives operators of enormous sums of money every year.. Fraud detection is an increasingly important and difficult task in today's technological environment. Several data mining applications are described and together they demonstrate that data mining can be used to identify telecommunication fraud, improve marketing effectiveness, and identify network faults. In this paper we propose a rule based for fraud detection in telecommunication system.

Continuous changes and the high calculation volume in network data distribution have made it more difficult to detect abnormal behaviors within and analyze data. For this cause, large data solutions have gained important. With the advancement of internet technologies and the digital age, cyber-attacks have increased steadily. The k-Means clustering algorithm is one of the most widely used algorithms in the world of data mining. Clustering algorithms are algorithms that automatically divide data into smaller clusters or subclusters. The algorithm places statistically similar records in the same group. In this article, we have used k-Means method from the Machine Learning libraries on Spark to determine whether the incoming network values are normal behavior. 400 thousand network data were used in this article. This data was obtained from KDD Cup 1999 Data. We have detected 10 abnormal behaviors from 400 thousand network data with k-means method.

International Journal of Electrical and Computer Engineering (IJECE), 2023

Fraud in healthcare insurance claims is one of the significant research challenges that affect the growth of the healthcare services. The healthcare frauds are happening through subscribers, companies and the providers. The development of a decision support is to automate the claim data from service provider and to offset the patient's challenges. In this paper, a novel hybridized big data and statistical machine learning technique, named MapReduce based iterative support vector machine (MR-ISVM) that provide a set of sophisticated steps for the automatic detection of fraudulent claims in the health insurance databases. The experimental results have proven that the MR-ISVM classifier outperforms better in classification and detection than other support vector machine (SVM) kernel classifiers. From the results, a positive impact seen in declining the computational time on processing the healthcare insurance claims without compromising the classification accuracy is achieved. The proposed MR-ISVM classifier achieves 87.73% accuracy than the linear (75.3%) and radial basis function (79.98%).

International Journal of Advanced Research in Science, Communication and Technology

Fraudulent transactions are very common problem in banking systems internationally, accounting for $5.1 trillion dollars every year. Many financial institutions are facing the common problem of being targeted by transactions of fraudulent nature and its becoming more and more obvious that advanced technology, such as Machine Learning (ML) is needed to counter such acts. Machine learning is the most effective technique against these complex bank frauds when approaches relying on fragmented and siloed data, rules-based approaches or traditional point-solutions are not only costly but also not as effective as needed. Complex algorithms powered by ML can be used to reduce manual investigations in Financial Institutions. Volume of these transactions is huge, lots of current solutions do not focus on big data the proposed model will work on big data with the help of ‘Apache Spark’ using latest machine learning technology. The proposed model will try to find pattern in given data set and f...

Workshop on Data …, 2006

Discovering Telecom Fraud Situations through Mining Anomalous Behavior Patterns Ronnie Alves, Pedro Ferreira, Orlando Belo, Joao Lopes, Joel Ribeiro University of Minho Campus de Gualtar 4710-057 Braga PORTUGAL {ronnie, pedrogabriel, obelo}@ di. uminho. pt Luís ...

2021

Noisy phone calls are aggravating and distracting, as well as frustrating. They may be classed as 'nuisance', 'emergency', 'random', and 'unsolicited' calls. Users have no inherent privileges on the internet; rather, their personalities are produced without any arrangement or evidence of involvement. It costs the U.S. communications company $8 billion per year to avoid call spam on the phone grid. Between January 2014 and June 2018, the FTC (Federal Trade Commission) received over 22 million reports of fraudulent and illegal telemarketing calls. Nowadays, the mobile network is used to issue automatic phone calls such as robocalls. Since it operates on text, we struggle with the following: What tactics and methods do we use to combat spam? Telephone TFD (Telecom Fraud Detection) here is discussed first. Concerning spam, we advanced our proposal by proposing a targeted traffic detection using a single weighted credibility algorithm with appropriate weig...

Journal of Computer Science & Computational Mathematics

Nowadays, adopting big data is a reality. Telecommunication company or telco must find the right solution to store all information available across the organization to maximize revenue using the analytics. The solution must be able to harness the large volume, variety, and velocity of the data available. One of the challenging actions is how to perform decision making and analysis in real-time. Some of the operational decisions may not comply with the corporation policy which makes it hard to keep up with the modern evolving business environment. Telco needs a platform to improve the business process and sustainable and profitable growth. The significant impacts should involve improvements of the customer experience and more reliable network quality, thereby reducing the customer churn rate. Big data and machine learning represent today's trends for the analytics. With big data analytics, the service provider can utilize the full potential of their data set by correlating, processing, and deciphering the hidden information from it. The conventional machine learning tools without big data are becoming inadequate as the trends shift towards distributed and real-time processing. The service provider needs the solution big data-driven which supports them to achieve timely manner and more accurate insights via the predictive analytics, text mining, and optimization. This paper also explains the characteristics of big data, and several uses of case implementing machine learning inside the big data platform related to telco operation such as mobile fraud detection. A well-known big data processing framework such as Hadoop indicated that there is an integration with machine learning tools such as Mahout, H2O.ai, R-Hadoop components, and KNIME. The advantages of these tools are evaluated based on their scalability, ease of use and extensibility features.

Fraud and default payments are two major anomalies in credit card transactions. Experts have been hard at work developing solutions, and one idea is to use data mining techniques. Credit card data, on the other side, can be tough for researchers to work with. This is related to the data qualities listed below: we have an unequal class distribution, and and redundant class samples. Both of these characteristics lead to low detection rates for minor anomalies in the data. Furthermore, defects in general learning algorithms contribute to the difficulty in categorising anomalies because the algorithms favour the majority class samples m. Each neural network configuration's findings are displayed and explained. To proposed advance big data classification technique based on fusion machine learning technique. The best method detects 98.81 percent of total fraud cases while minimising alarms by 30.32 percent.

EURAS Journal of Engineering and Applied Sciences, 2021

Artificial intelligence is used for many purposes nowadays. With the developments in technology, the fraudsters develop their methods. On the other hand, artificial intelligence methods are used in fraud detection for increasing the efficiency of corporations. AI and big data play an important role in real time data enrichment, deep learning integration and decisions. There are ten artificial intelligence methods explained which are used for fraud detection. Each method has its unique bases and it can not be said that there is only one optimal method. In this research, the methods are briefly explained, and a comparison is done for accuracy of methods. Supervised machine learning, unsupervised machine learning or semi-supervised machine learning as well as adaptive machine learning techniques against adaptive attacks with the advantage of big data and artificial intelligence are discussed with effectiveness usage for the future applications.