Analysis of third party networks
A method of analyzing customer behavior, where customers are engaged in customer-to-customer transactions in the third-party network, includes the transformation of data representing the customer-to-customer transactions from a data representation to a network representation, and then analyzing the network representation. The network representation includes a set of nodes and a set of links where each node represents a customer and each link represents a transaction between two of the customers.
1. A method of monitoring customer behavior comprising:
monitoring data corresponding to a plurality of individual transactions between customers;
transforming the data corresponding to a group of the plurality of individual transactions from a data representation into a network representation; and
analyzing the network representation to identify at least one transaction pattern within the network representation.
2. The method of claim 1 wherein the transforming step comprises building a network, wherein:
the network comprises a plurality of links and a plurality of nodes;
each node corresponds to a customer in a customer-to-customer transaction; and
each link corresponds to a transaction between two of the customers.
3. The method of claim 2 wherein the network has external network pattern properties, and the analyzing comprises analyzing internal link pattern properties and external network pattern properties.
4. The method of claim 3 wherein the external network properties show that the two customers are pan of a common external network.
5. The method of claim 2 wherein the transforming further comprises building a dictionary for the nodes.
6. The method of claim 2 wherein the analyzing treats all of the customers in the network as a single network object, and the transaction pattern comprises a group transaction pattern.
7. A storage medium containing computer program instructions that instruct a computing device to perform a method, the method comprising:
monitoring data corresponding to a plurality of individual transactions between customers;
transforming the data corresponding to a group of the plurality of individual transactions from a data representation into a network representation; and
analyzing the network representation to identify at least one transaction pattern within the network representation.
8. The storage medium of claim 7 wherein:
the method further comprises building a network;
the network comprises a plurality of links and a plurality of nodes;
each node corresponds to a customer in a customer-to-customer transaction;
each link corresponds to a transaction between two of the customers;
the network has external network pattern properties, and
the analyzing comprises analyzing internal link pattern properties and external network pattern properties.
9. The storage medium of claim 8 wherein the transforming further comprises building a dictionary for the nodes.
10. The storage medium of claim 9 wherein the external network properties show that the two customers are part of a common external network.
11. The storage medium of claim 8 wherein the analyzing treats all of the customers in the network as a single network object, and the transaction pattern comprises a group transaction pattern.
12. A method of monitoring customer behavior comprising:
monitoring data corresponding to a plurality of individual transactions between customers that occur at least partially in a network, wherein each transaction comprises a transfer of funds;
transforming the data corresponding to a group of the plurality of individual transactions from a data representation into a network representation wherein the network comprises a plurality of links and a plurality of nodes, each node corresponds to data indicative of a participant in one or more of the transactions, and each link is associated with two nodes and comprises data indicative of a relationship between the participants associated with the same two nodes; and
analyzing the network representation to identify at least one transaction pattern within the network representation.
13. The method of claim 12 comprising the additional step of detecting a link that corresponds to a pre-identified transaction pattern.