Use of abstracted data in pattern matching system

Abstract

Method, system and program product, for operating a call center system, the method comprising: obtaining performance data for agents in a set of agents; obtaining a respective abstracted data stream for multiple calls, each respective data stream having multiple different locations along the abstracted data stream representing multiple different respective fields, the meaning for the field data in the respective different locations for the different respective fields not known by the system; determining respective patterns for the respective data streams; obtaining performance data for the respective patterns; matching using a selected matching algorithm one of the agents from the set of agents to one of the calls based at least in part on the performance data for the respective pattern of the call and on performance data for the respective agents of the set of agents.

We claim: 
     
       1. A method of operating a call center system, comprising:
 obtaining, by one or more computers, performance data for agents in a set of agents;   obtaining, by the one or more computers, a respective abstracted data stream for each of multiple calls, with each respective data stream having multiple different locations along the abstracted data stream representing multiple different respective fields, wherein the meaning for the field data in the respective different locations for the different respective fields is not known by the system;   determining, by the one or more computers, from respective field data positions in the respective data streams respective patterns for the respective data streams;   obtaining, by the one or more computers, performance data for the respective patterns; and   matching, by the one or more computers, using a selected matching algorithm one of the agents from the set of agents to one of the calls based at least in part on the performance data for the respective pattern of the call and based at least in part on the performance data for the respective agents of the set of agents.   
     
     
       2. The method as defined in  claim 1 , wherein the abstracted data stream is one selected from the group of 0's and 1's, yes and no, true and false, binary, integer, decimal, and alphanumeric. 
     
     
       3. The method as defined in  claim 1 , wherein the abstracted data stream is scrambled, but the scrambling is the same for calls received during a period. 
     
     
       4. The method as defined in  claim 1 , wherein the abstracted data stream is scrambled, and further comprising:
 rearranging, by the one or more computers, the data to an order of a data set to be used for correlation.   
     
     
       5. The method as defined in  claim 1 , further comprising the steps:
 determining, by the one or more computers, field positions for the fields in the abstracted data stream; and   selecting, by the one or more computers, only the determined field positions from the data stream to form the pattern.   
     
     
       6. The method as defined in  claim 1 ,
 wherein for multiple of the abstracted data streams received during the period, the abstracted data for the different fields are received in two or more different orders over the course of the period of time, and   further comprising arranging, by the one or more computers, the field data for the different fields for the data streams of the respective calls in a same order, by the one or more computers.   
     
     
       7. The method as defined in  claim 1 , wherein the respective pattern determined for the call is ranked by performance, and wherein the matching algorithm performs matching based at least in part on the performance rank of the call and the performance data of the respective agents in the set of agents. 
     
     
       8. The method as defined in  claim 1 , wherein the performance of the call is based on one or more selected from the group of sale, call handle time, retention, customer satisfaction, revenue, first call resolution, units sold, and transaction points. 
     
     
       9. The method as defined in  claim 1 , further comprising:
 obtaining, by the one or more computers, a call type for the respective call selected from the group of sale, complaint, information request, and very important person (VIP), and   wherein the matching algorithm performs matching based at least in part on the call type.   
     
     
       10. The method as defined in  claim 1 , wherein the matching step during a period further comprises:
 switching, by the one or more computers, to using a different matching algorithm based on correlation of call performance to agent performance at a switchover point in the period.   
     
     
       11. A system of operating a call center system, comprising:
 a communications component for handling calls; and   one or more computers configured with the following components;   a collector configured in the one or more computers to obtain performance data for agents in a set of agents;   the collector configured in the one or more computers to obtain a respective abstracted data stream for each of multiple of the calls, with each respective data stream having multiple different locations along the abstracted data stream representing multiple different respective fields, wherein the meaning for the field data in the respective different locations for the different respective fields is not known by the system;   the collector configured in the one or more computers to determine from respective field data positions in the respective data streams respective patterns for the respective data streams;   the collector configured in the one or more computers to obtain performance data for the respective patterns; and   a matching engine configured in the one or more computers with a matching algorithm to match one of the agents from the set of agents to one of the calls based at least in part on the performance data for the respective pattern of the call and based at least in part on the performance data for the respective agents of the set of agents.   
     
     
       12. The system as defined in  claim 11 , wherein the abstracted data stream is one selected from the group of 0's and 1's, yes and no, true and false, binary, integer, decimal, and alphanumeric. 
     
     
       13. The system as defined in  claim 11 ,
 wherein abstracted data stream is scrambled, and   wherein the collector is configured in the one or more computers to rearrange the data to an order of a data set to be used for correlation.   
     
     
       14. The system as defined in  claim 11 ,
 wherein the collector is configured in the one or more computers   to determine field positions for the fields in the abstracted data stream; and   to select only the determined field positions from the data stream to form the pattern.   
     
     
       15. The system as defined in  claim 11 ,
 wherein for multiple of the abstracted data streams received during the period, the abstracted data for the different fields are received in two or more different orders over the course of the period of time, and   wherein the collector is further configured in the one or more computers to arrange the field data for the different fields for the data streams of the respective calls in a same order.   
     
     
       16. The system as defined in  claim 11 ,
 wherein the matching engine is configured to rank the respective patterns by performance, and to perform matching based at least in part on the performance rank of the call and the performance data of the respective agents in the set of agents.   
     
     
       17. Non-transitory computer-readable media, comprising:
 computer-readable program code embodied therein, to perform, when executed by one or more computers, the steps:   obtaining, by the one or more computers, performance data for agents in a set of agents;   obtaining, by the one or more computers, a respective abstracted data stream for each of multiple calls, with each respective data stream having multiple different locations along the abstracted data stream representing multiple different respective fields, wherein the meaning for the field data in the respective different locations for the different respective fields is not known by the system;   determining, by the one or more computers, from respective field data positions in the respective data streams respective patterns for the respective data streams;   obtaining, by the one or more computers, performance data for the respective patterns; and   matching, by the one or more computers, using a selected matching algorithm one of the agents from the set of agents to one of the calls based at least in part on the performance data for the respective pattern of the call and based at least in part on the performance data for the respective agents of the set of agents.   
     
     
       18. A method for pairing in a contact center system comprising:
 receiving, by at least one computer processor communicatively coupled to and configured to operate in the contact center system, abstracted training data for a plurality of contacts or a plurality of agents;   generating, by the at least one computer processor, a computer-based model for optimizing contact-agent pairs in the contact center system using the abstracted training data; and   establishing, by the at least one computer processor, in a switch module of the contact center system, a communication channel between a first contact of the plurality of contacts and a first agent of the plurality of agents using the computer-based model to optimize performance of the contact center system.   
     
     
       19. The method of claim 18, wherein the abstracted training data is not encrypted, and wherein the abstracted training data cannot be decrypted by the at least one computer processor or the contact center system. 
     
     
       20. The method of claim 18, wherein the abstracted training data includes customer relationship management (CRM) data. 
     
     
       21. The method of claim 18, wherein the abstracted training data includes sensitive information. 
     
     
       22. The method of claim 18, wherein the abstracted training data includes identifying personal information. 
     
     
       23. The method of claim 18, wherein establishing the communication channel comprises:
 selecting, by the at least one computer processor, a pairing of the first contact with the first agent based on abstracted data of the first contact or the first agent correlated with the abstracted training data in the generated computer-based model.   
     
     
       24. The method of claim 23, wherein the selected pairing has a worse expected outcome than a pairing of the first contact with a second agent of the plurality of agents or the first agent with a second contact of the plurality of contacts. 
     
     
       25. The method of claim 1, further comprising:
 cycling, by the at least computer processor, selection using the computer-based model on and off; and   measuring overall performance of the computer-based model in the contact center system by comparing a first actual performance of the contact center system while the computer-based model is cycled on with a second actual performance of the contact center system while the computer-based model is cycled off.   
     
     
       26. A system for pairing in a contact center system comprising:
 at least one computer processor communicatively coupled to and configured to operate in the contact center system, wherein the at least one computer processor is further configured to:
 receive abstracted training data for a plurality of contacts or a plurality of agents; 
 generate a computer-based model for optimizing contact-agent pairs in the contact center system using the abstracted training data; and 
 establish, in a switch module of the contact center system, a communication channel between a first contact of the plurality of contacts and a first agent of the plurality of agents using the computer-based model to optimize performance of the contact center system. 
   
     
     
       27. The system of claim 26, wherein the abstracted training data is not encrypted, and wherein the abstracted training data cannot be decrypted by the at least one computer processor or the contact center system. 
     
     
       28. The system of claim 26, wherein the abstracted training data includes customer relationship management (CRM) data. 
     
     
       29. The system of claim 26, wherein the abstracted training data includes sensitive information. 
     
     
       30. The system of claim 26, wherein the abstracted training data includes identifying personal information. 
     
     
       31. The system of claim 26, wherein establishing the communication channel comprises selecting a pairing of the first contact with the first agent based on abstracted data of the first contact or the first agent correlated with the abstracted training data in the generated computer-based model. 
     
     
       32. The system of claim 31, wherein the selected pairing has a worse expected outcome than a pairing of the first contact with a second agent of the plurality of agents or the first agent with a second contact of the plurality of contacts. 
     
     
       33. The system of claim 26, wherein the at least one computer processor is further configured to:
 cycle selection using the computer-based model on and off; and   measure overall performance of the computer-based model in the contact center system by comparing a first actual performance of the contact center system while the computer-based model is cycled on with a second actual performance of the contact center system while the computer-based model is cycled off.   
     
     
       34. An article of manufacture for pairing in a contact center system comprising:
 a non-transitory processor readable medium; and   instructions stored on the medium;   wherein the instructions are configured to be readable from the medium by at least one computer processor communicatively coupled to and configured to operate in the contact center system and thereby cause the at least one computer processor to operate so as to:
 receive abstracted training data for a plurality of contacts or a plurality of agents; 
 generate a computer-based model for optimizing contact-agent pairs in the contact center system using the abstracted training data; and 
 establish, in a switch module of the contact center system, a communication channel between a first contact of the plurality of contacts and a first agent of the plurality of agents using the computer-based model to optimize performance of the contact center system. 
   
     
     
       35. The article of manufacture of claim 34, wherein the abstracted training data is not encrypted, and wherein the abstracted training data cannot be decrypted by the at least one computer processor or the contact center system. 
     
     
       36. The article of manufacture of claim 34, wherein the abstracted training data includes customer relationship management (CRM) data. 
     
     
       37. The article of manufacture of claim 34, wherein the abstracted training data includes sensitive information. 
     
     
       38. The article of manufacture of claim 34, wherein the abstracted training data includes identifying personal information. 
     
     
       39. The article of manufacture of claim 34, wherein establishing the communication channel comprises selecting a pairing of the first contact with the first agent based on abstracted data of the first contact or the first agent correlated with the abstracted training data in the generated computer-based model. 
     
     
       40. The article of manufacture of claim 39, wherein the selected pairing has a worse expected outcome than a pairing of the first contact with a second agent of the plurality of agents or the first agent with a second contact of the plurality of contacts. 
     
     
       41. The article of manufacture of claim 34, wherein the at least one computer processor is further caused to operate so as to:
 cycle selection using the computer-based model on and off; and   measure overall performance of the computer-based model in the contact center system by comparing a first actual performance of the contact center system while the computer-based model is cycled on with a second actual performance of the contact center system while the computer-based model is cycled off.