Carrying out predictive analysis relating to nodes of a communication network

• US 8,644,468 B2
• Filed: 08/26/2011
• Issued: 02/04/2014
• Est. Priority Date: 12/15/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dantressangle, Patrick, Hechler, Eberhard, Oberhofer, Martin, Wurst, Michael
• Primary Examiner(s): Moore, Ian N
• Assistant Examiner(s): Smith, Eric P
• Application Number:
  US13/219,362
• Publication Number:
  US 20120155290A1
• Time in Patent Office: 893 Days
• Field of Search: 705/7.31, 707/776, 707/777, 707/778
• US Class Current: 379/114.01
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/5064 Customer relationship management

Abstract

Predictive analysis relating to nodes of a communication network is carried out by providing communication event information for a first set of nodes and a second set of nodes of the communication network, providing a set of attributes for the nodes of the first set, using the attributes and the communication event information for determining a set of groups among the first set of nodes, assigning each node of the second set to at least one group of the set of groups based at least on the communication event information available for the second group, the assigning resulting in membership information of the nodes of the second set, and deriving or applying a prediction model for the second set of nodes based on the communication event information for the second set and the membership information.

Claims

1. A method for carrying out predictive analysis relating to nodes of a communication network, comprising:

providing, using a processor of a computer, communication event information for a first set of nodes and a second set of nodes of the communication network;

providing a set of attributes for the nodes of the first set of nodes, wherein attributes are not available for the nodes of the second set of nodes;

using the set of attributes and the communication event information to determine a set of groups among the first set of nodes;

assigning each node of the second set of nodes to at least one group of the set of groups based at least on the communication event information for the second set of nodes, wherein the assigning results in membership information of the nodes of the second set of nodes, and wherein the membership information becomes a first attribute for each node of the second set of nodes;

assigning at least one membership weight to each node of the second set of nodes, wherein the weight becomes a second attribute for each node of the second set of nodes; and

deriving a prediction model for the second set of nodes based at least on the communication event information for the second set of nodes, the first attribute of membership information, and the second attribute of weight.

2. The method according to claim 1, wherein a node from the second set of nodes that is assigned to more than one group is added to each group with a fraction of the weight assigned to the node.

3. The method according to claim 1, wherein the assigning of each node of the second set of nodes is done based on the communication event information available for the first set of nodes and the second set of nodes.

4. The method according to claim 1, wherein the assigning of each node of the second set of nodes comprises:

calculating an aggregated distribution for each group of nodes of the first set of nodes; and

using a statistical measure to compare the communication event information of the nodes of the second set of nodes to the aggregated distribution of each group.

5. The method according to claim 1, wherein the assigning of each node of the second set of nodes further comprises:

describing the communication event information with at least one of a duration of service usage, a distribution over time, and a geographical distribution.

6. The method according to claim 1, wherein the first set of nodes relates to information-rich nodes and the second set of nodes relates to information-poor nodes.

7. The method according to claim 1, wherein the assigning of each node of the second set of nodes further comprises:

assigning the nodes of the second set of nodes to one or more subscriber segmentation groups based on at least one of general call patterns, temporal call patterns, geographic patterns, communication patterns, and service usage patterns.

8. The method according to claim 1, further comprising:

using the predictive analysis to perform at least one of comparing network/call-data-records-derived insight and service usage patterns against globally available demographic data and comparing network/call-data-records-derived insight and service usage patterns against existing post-paid subscriber insight and demographic data from any specific Telecommunication Service Provider.

9. The method according to claim 1, wherein the communication event information includes communication patterns.

10. The method according to claim 1, wherein the prediction model is derived by training.

11. The method according to claim 10, wherein the training is performed by finding a statistical model that associates one or more variables representing a purpose of doing predictive analysis with a set of properties of a node of the second set.

12. A computer program product for carrying out predictive analysis relating to nodes of a communication network, the computer program product comprising:

a computer-readable storage device storing a computer readable program, wherein the computer readable program, when executed on a computer, performs;

providing communication event information for a first set of nodes and a second set of nodes of the communication network;

providing a set of attributes for the nodes of the first set of nodes, wherein attributes are not available for the nodes of the second set of nodes;

using the set of attributes and the communication event information to determine a set of groups among the first set of nodes;

assigning each node of the second set of nodes to at least one group of the set of groups based at least on the communication event information for the second set of nodes, wherein the assigning results in membership information of the nodes of the second set of nodes, and wherein the membership information becomes a first attribute for each node of the second set of nodes;

assigning at least one membership weight to each node of the second set of nodes, wherein the weight becomes a second attribute for each node of the second set of nodes; and

deriving a prediction model for the second set of nodes based at least on the communication event information for the second set of nodes, the first attribute of membership information, and the second attribute of weight.

13. The computer program product according to claim 12, wherein a node from the second set of nodes that is assigned to more than one group is added to each group with a fraction of the weight assigned to the node.

14. The computer program product according to claim 13, wherein the computer readable program, when executed on the computer, performs:

using the at least one membership weight of each node in the second set of nodes as additional features for the prediction model.

15. The computer program product according to claim 12, wherein the assigning of each node of the second set of nodes is done based on the communication event information available for the first set of nodes and the second set of nodes.

16. A data processing system for carrying out predictive analysis relating to nodes of a communication network, comprising:

a processor; and

a storage device connected to the processor, wherein the storage device has stored thereon a program, and wherein the processor is configured to execute the program to perform operations, wherein the operations comprise;

providing communication event information for a first set of nodes and a second set of nodes of the communication network;

providing a set of attributes for the nodes of the first set of nodes, wherein attributes are not available for the nodes of the second set of nodes;

using the set of attributes and the communication event information to determine a set of groups among the first set of nodes;

assigning each node of the second set of nodes to at least one group of the set of groups based at least on the communication event information for the second set of nodes, wherein the assigning results in membership information of the nodes of the second set of nodes and wherein the membership information becomes a first attribute for each node of the second set of nodes;

assigning at least one membership weight to each node of the second set of nodes, wherein the weight becomes a second attribute for each node of the second set of nodes; and

deriving a prediction model for the second set of nodes based at least on the communication event information for the second set of nodes, the first attribute of membership information, and the second attribute of weight.

17. The data processing system according to claim 16, wherein a node from the second set of nodes that is assigned to more than one group is added to each group with a fraction of the weight assigned to the node.

18. The data processing system according to claim 17, wherein the computer readable program, when executed on the computer, performs:

using the at least one membership weight of each node in the second set of nodes as additional features for the prediction model.

19. The data processing system according to claim 16, wherein the assigning of each node of the second set of nodes is done based on the communication event information available for the first set of nodes and the second set of nodes.