KARACHI: In recent times, the government has resorted to extreme measures of blocking mobile phone SIMs to thwart potential terrorist attacks. This cannot be a long-term solution to the terror problem. Instead, what we need is an analytical framework that aids competent authorities in identifying suspicious use of mobile phone SIMs.
Artificial intelligence (AI)-based techniques can help identify mobile phone SIMs that are being used for illegal activities. We already have precedence in the form of the Anti-money Laundering Act that requires banks to report suspicious activities to a financial monitoring unit that may initiate further investigation against the beneficiary or executor of a suspicious transaction. To comply with the Act, all banks have deployed AI and statistical reasoning-based tools that help in the identification of suspicious patterns. In a similar manner, automated tools that can identify suspicious usage patterns of mobile phone SIMs need to be deployed by mobile network operators and their findings should be reported to the Pakistan Telecommunication Authority, the interior ministry and other relevant authorities.
AI techniques have been used the world over by the telecommunication industry to develop churn models that help mobile network operators predict customer attrition. One can deploy similar AI techniques to predict if a particular mobile phone SIM is being used in a suspicious way. It can also be argued that learning from the experience of the banking sector, in tackling suspicious activities, would also be useful in developing the proposed analytical framework. For instance, if a person does not use his bank account for six months, then the account becomes dormant. In a similar way, a mobile phone SIM not being used for a few months should be deactivated and a personal visit should be the only mechanism for its reactivation.
There is no doubt that a successful implementation of such an analytical framework would require research and development efforts by both academia and industry but it could be a viable long-term solution to the security situation.
Dr Sajjad Haider
Associate Professor, Faculty of Computer Science, IBA
Published in The Express Tribune, November 17th, 2012.
