Systems and methods for generating leads in a network by predicting properties of external nodes

Abstract

The present invention is directed towards systems and methods for predicting one or more desired properties of external nodes or properties of their relations with internal nodes, based on a selected group of nodes about which it is known whether the nodes have the desired properties, or it is known whether they have a desired relation property with an internal node. The method comprises storing in one or more data structures a first data set regarding external nodes and a second data set regarding nodes with known properties in a selected group, each data set having one or more data items representing one or more events relating to or attributes of each node in the data set, the second data set including one or more types of data items not included in the first data set. The method then models the second data set to identify from the second data one or more modeled events or attributes of internal nodes in the selected group that are statistically likely to identify the nodes or their relations, that have the desired properties and predicts which of the external nodes are statistically likely to have the one or more desired properties, or desired relation property with internal node, based on the identified plurality of modeled events or attributes and the events or attributes in the first data set.

1 . A computerized method for predicting one or more desired properties of external nodes based on a selected group of nodes about which it is known whether the nodes have the one or more desired properties, wherein the one or more desired properties are not known with respect to the one or more external node because the one or more external nodes lack a relationship with a given entity, the method comprising:
 storing in one or more data structures a first data set regarding external nodes and a second data set regarding nodes in a selected group, each data set having one or more data items representing one or more events relating to or attributes of each node in the data set, the second data set including one or more types of data items not included in the first data set;   virtualizing the second data set regarding nodes into a modeled second data set after the first data set regarding external nodes at least by eliminating from the second data set the one or more data item types not included in the first data set;   modeling the virtualized second data set to identify from the modeled second data one or more modeled events or attributes of nodes in the selected group that are statistically likely to identify the nodes that have the desired properties; and   predicting which of the external nodes are statistically likely to have the one or more desired properties based on the identified plurality of modeled events or attributes and the events or attributes in the first data set.   
     
     
         2 . The method of  claim 1 , wherein storing the second data set comprises storing data items regarding the selected group of nodes that all are known to have the one or more desired properties. 
     
     
         3 . The method of  claim 1 , wherein storing the second data set comprises storing data items regarding a positive set of nodes that are known to have the one or more desired properties and a negative set of nodes that are known to not have the one or more desired properties. 
     
     
         4 . The method of  claim 1 , wherein storing the second data set comprises storing data items regarding a weighted set of nodes, the weighted set comprising a set of one or more desired properties. 
     
     
         5 . The method of  claim 3 , wherein modeling the modeled second data set comprises identifying from the modeled second data set one or more first collections of events or attributes of nodes in the selected group that are statistically likely to make the nodes in the selected group that have the desired properties and one or more second collections of events or attributes of nodes in the selected group that are statistically likely to identify the nodes in the selected group that do not have the desired properties. 
     
     
         6 . The method of  claim 5 , wherein modeling the modeled second data set comprises identifying from the modeled second data set events or attributes that generate a statistically high distinction between the nodes in the positive set and the nodes in the negative set. 
     
     
         7 . The method of  claim 5 , wherein statistically analyzing the modeled second data set comprises executing a machine learning algorithm program. 
     
     
         8 . The method of  claim 1 , comprising generating the second data set from a larger data set. 
     
     
         9 . The method of  claim 8 , wherein generating the second data set comprises selecting nodes from the larger data set based on one or more data items in the larger set having values identifying the corresponding nodes as having or not having the one or more desired properties. 
     
     
         10 . The method of  claim 1 , wherein the selected group of nodes represent customers of a service provider and the external nodes represent entities that are not customers of the service provider, wherein the one or more desired properties comprise one or more properties of the non-customer entities that are specified by the service provider, and wherein storing the first and second data sets comprises storing data received from the service provider. 
     
     
         11 . The method of  claim 10 , wherein storing the first and second data sets comprises storing transaction data regarding transactions involving customer and non-customer entities of the service provider. 
     
     
         12 . The method of  claim 11 , wherein storing transaction data comprises storing transaction data for the second data set including one or more data items related to transactions involving the service provider customers, the data items only being receivable for transactions involving the service provider customers. 
     
     
         13 . The method of  claim 12 , wherein the one or more data items related to transactions comprise transactions involving only service provider customers. 
     
     
         14 . The method of  claim 11 , wherein the service provider comprises a telephone operator, and wherein storing transaction data comprises storing call detail records (CDRs) of telephone calls by or to customers and non-customer entities. 
     
     
         15 . The method of  claim 11 , wherein the service provider comprises an e-mail provider, and wherein storing transaction data comprises storing e-mail records by or to users and non-users of the e-mail provider. 
     
     
         16 . The method of  claim 15 , wherein storing e-mail records comprises storing one or more e-mail record data items selected from the group consisting of: e-mail addresses, subject lines of e-mails, dates of e-mails, the number of e-mails sent or received, and a contact list associated with an e-mail. 
     
     
         17 . The method of  claim 16 , wherein the one or more desired properties comprises properties of a given entity selected from the group consisting of: volume of emails generated by the entity, friends of the entity that are non-users of the e-mail provider, social influence of a non-user of the email provider, and the entity having certain gender, age group, location, or occupation. 
     
     
         18 . The method of  claim 11 , wherein the service provider comprises a social network operator, and wherein storing transaction data comprises storing records of relations or interaction between users and non-users of the social network provider. 
     
     
         19 . The method of  claim 18 , wherein the one or more desired properties comprise properties of entities selected from the group consisting of: social influence of a given user, number of connections a given user has, age of a given user, location of a given user, amount of time spent at the social network, gender of a given user, and the given user's interest. 
     
     
         20 . The method of  claim 11 , wherein the service provider comprises a delivery service provider, and wherein storing transaction data comprises storing delivery records between customers and non-customers of the delivery service provider. 
     
     
         21 - 55 . (canceled)