Telecommunications Telecommunications fraud includes a wide range of deceptions used by perpetrators to receive telephony services free of charge or at a reduced rate. TYPES OF FRAUD There are many methods employed by unscrupulous elements to defraud telecom networks. Subscription fraud Subscription fraud is the most damaging among the non-technical frauds. The methodology adopted here is simple, with the fraudster obtaining a connection from the telecom company using the normal, accepted procedure of the company. The fraudster has no intention to pay and rings up huge bills in very short time frames. Cloning Fraud Cloning occurs when a customer's MIN and ESN are programmed into a cellular telephone not belonging to the customer. Cloning Fraud through Tumbling A tumbler phone contains a special chip that rotates through numerous legitimate and false ESN/MIN pairs. Call-Sell Fraud In call-sell fraud the fraudster abuses the call forwarding feature of networks and uses his connection to set-up multiple, simultaneous calls for his “clients.” Premium Rate Service Fraud The PRS content provider obtains a subscription and then uses this subscription to make calls to their own PRS service and thus inflate the transactions to their service. Surfing Fraud Surfing involves the use of another person’s service without having the necessary authority. Ghosting Fraud Deceiving the network in order to obtain free or cheap rate calls. Accounting Fraud Accounting involves exploiting accounting and billing processes to reduce charges or obtain a cash refund. This method usually involve “insiders” in some way. Roaming Fraud Nearly all mobile phones allow roaming between networks and across countries. Fraudsters exploit the time lag in sharing account activation and billing data between these operators. Handset Theft Handsets are an attractive target for theft. In the window of time before the theft is reported and the account barred, many high value calls can be placed. Stolen phones are also frequently cloned. Pre-paid Systems Fraudsters have found it possible to defraud pre-paid platforms through the abuse of security codes, using ghosting techniques to confuse the platform. Internal Fraud In addition to the above, is the problem of internal fraud wherein employees within the operator or associates of the company aid outsiders in defrauding the network. Employees have access to all parts of the network and are sufficiently knowledgeable to know where to play the dirty tricks. CURRENT SOLUTIONS To date, operators have primarily relied on “threshold based” analysis, “rules-based” systems or “neural networks.” These systems detect fraud by monitoring individual and aggregated calls for indicators such as overlap, very high value, and very long duration calls, hence indicating that a potential fraud is under way. These systems have not, however, always been successful. Key trigger characteristics can occur in both fraudulent and non-fraudulent calls, and to avoid inundating human analysts with false alarms, operators have often de-tuned such systems to generate an acceptable workload. Inevitably, more frauds have slipped through the coarser net. TEN POWERFUL ARTIFICIAL INTELLIGENCE TECHNOLOGIES TO BUILD A BETTER SOLUTION Ten powerful artificial intelligence techniques will stop more, and stop more subtle fraud than will one or two techniques alone. Each technology is effective at expressing a different type of relationship among the attributes of a transaction and relevant data. Brighterion has optimized each of its ten technologies for performance in real-time operation, and integrated them into a single software solution. Multi-agent technology breaks through limitations of algorithmic techniques and enables “first fraud” capability Purely algorithmic techniques have a serious limitation: they can only identify the specific patterns they have been trained to recognize. In multi-agent technology, by contrast, multiple independent agents are each assigned one or more goals, then interact and negotiate with each other in order to reach their individual and collective goals. Brighterion is a pioneer in the development of goal-based agent technology, and is the first company to apply it to a commercial anti-fraud product. iPrevent is uniquely capable of identifying “abnormal behavior” and therefore the only product on the market with the ability to identify, and stop, that most elusive and costly “first fraud.” BENEFITS OF IPREVENT FOR TELECOM FRAUD PREVENTION Real time detection of fraudulent activity Immediate rating and segmentation of records Continuous user profiling and aberration alarms Dynamic management of fraud “Hot Lists” Automated pattern analysis to find “Points of Administration” “Fingerprinting” of known fraud occurrences Credit thresholds and alarms Consolidation of multiple risk events Advanced case and work process management Both rigid and “fuzzy” linking between cases A GUI to enable flag settings specific to each situation or customer iPrevent LiveLog, to display each of the records that have been referred to a case investigator Information about the customer (eg..“date” “time” “cost,” etc..) Scalability, through deployment of clusters of server farms