Enhancing Fraud Detection with Privacy-Preserving Record Linkage and Digital Signature (EdDSA) Techniques: A Data Sharing Approach

– Fraud as a threat to the telecom business is very real amid escalating and hopeless problem. Many telecoms also have different sets of measures that they employ against fraud. The prevention of fraud requires information and it is through this premise that detection comes in. There is also the lack of privacy rights for the exchange of information as people’s privacy becomes restricted. There are a variety of techniques defined to ensure that data sharing can also include PPRL considerations. Current works are effective due to the reasons that many of the PPRL techniques adopt a similarity measure which could be Jacquard similarity on relevant datasets. As it has been mentioned earlier, the Bloom filter implementation is applied in a number of complex and slow algorithms; however, the given technique is critically sensitive to the cryptanalysis attacks. Based on current telco infrastructure and without having to go through a multistep protocol mechanism, this paper introduces a PPRL way which is indeed invulnerable to attacks. Starting with a fresh approach to matching non-homologous datasets is the application of the new attack-proof Digital signature system like the Edwards curve. Note that what is capability of this approach is only to estimate the Jaccard similarity here, which the datasets cannot be used. It’s also important to note that basic request-response model is implemented for two-partner interactions. The performance, privacy, and matching accuracy of the approach have been evaluated by employing a scale of 1 to 5 using a large set of data set public. Concrete against attacks, this technique provides an enhanced pace and attains perfect match consistency.