| 摘要 |
In one embodiment, the present invention includes methods for detecting fraudulent or erroneous invoices that use an associative memory and models of valid invoices. Incoming invoices may be applied against the models of valid invoices, and the models may access the associations of data elements in an associative memory and provide results to a decision engine. The decision engine may flag invoices that are nonconforming for further analysis. As more invoices and information about the invoices are gathered, the models and associative memory may be adapted to improve detection performance. |
