Methods and systems for providing automated predictive analysis

Abstract

A method for providing automated predictive analysis for a case management system is disclosed and includes providing, by a server, a case management system comprising configuration information defining a plurality of case attributes and performance criteria comprising at least one performance criterion, and comprising case management data associated with the plurality of case attributes and a plurality of cases, and automatically determining a projected outcome associated with a performance criterion included in the configuration information. The method also includes identifying a correlation between a case attribute and the projected outcome based on case management data associated with the case attribute and/or case management data associated with the performance criterion, generating a prediction model based on the correlation and providing a prediction relating to the projected outcome based on the correlation.

What is claimed is: 
     
         1 . A system comprising:
 a case management server coupled by an application protocol layer to a network subsystem, the case management server including a self-learning predictive analysis engine and an outgoing data handler that interoperates directly with the application protocol layer;   a data store coupled to the case management server;   a first remote computer device communicatively coupled to the case management server by the network subsystem;   wherein the case management server defines case management data associated with a plurality of cases, the defined case management data including user-configured case attributes and user-configured performance criteria, the performance criteria defining predetermined goals for associated ones of the case attributes;   wherein the data store contains configuration data including the defined case management data;   wherein the self-learning predictive analysis engine is configured to:
 analyze a first portion of the configuration data containing a plurality of the case attributes and a corresponding plurality of the performance criteria and associated outcomes in a first portion of the configuration data and generate a set of projected outcomes corresponding to the performance criteria, 
 generate correlation data that establishes, for each of the case attributes, a relationship with a corresponding one of the projected outcomes based on at least one of the performance criteria corresponding to respective ones of the case attributes and the projected outcomes, and 
 store the correlation data in the data store; 
   wherein the self-learning predictive analysis engine is further configured to thereafter;
 receive case data including one or more attributes for a corresponding specific case from the first remote computer device, 
 identify values of the one or more attributes in the received case data, 
 generate, from the stored correlation data, a first prediction model using a first predictive algorithm and a second prediction model using a second prediction algorithm which is different than the first prediction algorithm, 
 test the first and second prediction models with an actual outcome to select one prediction model that best represents the defined case management data, and 
 identify, using the selected prediction model, for the corresponding specific case, ones of the projected outcomes in the stored correlation data that are associated with the identified case attribute values in the received case data; 
   wherein the case management server is configured to:
 build a message based on the identified projected outcomes for the corresponding specific case, and 
 provide the message to the application protocol layer; 
   wherein the application protocol layer is configured to format the received message for delivery to the first remote computer device.   
     
     
         2 . The system of  claim 1 , wherein the self-learning predictive analysis engine is configured to:
 generate the correlation data based on the first portion of the configuration data;   retrieve a second portion of the configuration data which is different from the first portion of the configuration data; and   for each of one or more cases in the second portion of the configuration data: identify attributes in the case data: identify in the stored correlation data specific projected outcomes associated with the identified case attributes, compare the identified specific projected outcomes with respective outcomes identified for the plurality of cases in the configuration data, and verify the specific projected outcomes with the respectively identified outcomes for the plurality of cases in the configuration data.   
     
     
         3 . The system of  claim 1 , further comprising an incoming message handler configured to interface with the application protocol layer and to receive the case data from the first remote computer device. 
     
     
         4 . The system of  claim 3 , further comprising a command handler coupled to the incoming message handler and configured to route the configuration data and the case data received via the incoming message handler. 
     
     
         5 . The system of  claim 4 , wherein when the command handler receives incoming configuration data, the command handler is configured to route the incoming configuration data to a data manager of the case management server, wherein the data manager is configured to store the received configuration data in the data store. 
     
     
         6 . The system of  claim 4 , wherein when the command handler receives incoming case data from the first remote computer device, the command handler is configured to route the incoming case data to the self-learning predictive analysis engine. 
     
     
         7 . The system of  claim 1 , wherein the self-learning predictive analysis engine is further configured to validate the generated correlation data against the received case data, update the generated correlation data according to the received case data, and store the updated correlation data. 
     
     
         8 . A computer program product, comprising a non-transitory computer readable storage medium that stores thereon computer-readable program code that is executable by a processor to:
 execute a case management system configured to build a message for delivery via an application protocol layer coupled to a network subsystem to a first remote computer device, the case management system including a self-learning predictive analysis engine and an outgoing data handler interoperating directly with the application protocol layer, the case management system defining case management data associated with a plurality of cases, the defined case management data including user-configured case attributes and user-configured performance criteria, the performance criteria defining predetermined goals for associated ones of the case attributes;   store, in a data store, configuration data including the defined case management data;   analyze, by the self-learning predictive analysis engine, a first portion of the configuration data containing a plurality of the case attributes and a corresponding plurality of the performance criteria and associated outcomes in a first portion of the configuration data and generate a set of projected outcomes corresponding to the performance criteria;   generate, by the self-learning predictive analysis engine, correlation data which establishes, for each of the case attributes, a relationship with a corresponding one of the projected outcomes based on at least one of the performance criteria corresponding to respective ones of the case attributes and the projected outcomes;   store, in the data store, the correlation data;   receive, by the self-learning predictive analysis engine, case data including one or more attributes for a corresponding specific case from the first remote computer device via the network subsystem;   identify, by the self-learning predictive analysis engine, values of the one or more attributes in the received case data;   generate, from the stored correlation data, a first prediction model using a first predictive algorithm and a second prediction model using a second prediction algorithm which is different than the first prediction algorithm;   test the first and second prediction models with an actual outcome to select one prediction model that best represents the defined case management data;   identify, by the self-learning predictive analysis engine using the selected prediction model, for the corresponding specific case, ones of the projected outcomes in the stored correlation data that are associated with the identified case attributes values in the received case data;   build, by the outgoing data handler, the message based on the identified projected outcomes for the corresponding specific case; and   provide, by the outgoing data handler, the message to the application protocol layer;   format the received message for delivery to the first remote computer device.   
     
     
         9 . The computer program product of  claim 8 , wherein the program code is further executable by the processor to:
 generate, by the case management system, the correlation data based on the first portion of the configuration data;   retrieve, by the case management system, a second portion of the configuration data which is different from the first portion of the configuration data; and   for each of one or more cases in the second portion of the configuration data: identify, by the case management system, attributes in the case data, identify in the stored correlation data specific projected outcomes associated with the identified case attributes, compare the identified specific projected outcomes with respective outcomes identified for the plurality of cases in the configuration data, and verify the specific projected outcomes with the respectively identified outcomes for the plurality of cases in the configuration data.   
     
     
         10 . The computer program product of  claim 8 , wherein the program code is further executable by the processor to implement an incoming message handler which interfaces with the application protocol layer and receives the case data from the first remote computer device. 
     
     
         11 . The computer program product of  claim 10 , wherein the program code is further executable by the processor to implement a command handler which is coupled to the incoming message handler and configured to route the configuration data and the case data received via the incoming message handler. 
     
     
         12 . The computer program product of  claim 11 , wherein the program code is further executable by the processor to, when the command handler receives incoming configuration data, route the incoming configuration data to a data manager that stores the received configuration data in the data store. 
     
     
         13 . The computer program product of  claim 11 , wherein the program code is further executable by the processor to, when the command handler receives incoming case data from the first remote computer device, route the incoming case data to the self-learning predictive analysis engine. 
     
     
         14 . The computer program product of  claim 8 , wherein the program code is further executable by the processor to validate, by the self-learning predictive analysis engine, the generated correlation data against the received case data, update the generated correlation data according to the received case data, and store the updated correlation data. 
     
     
         15 . A method executed in a case management system having a self-learning predictive analysis engine and an outgoing data handler, the method comprising:
 executing the case management system to build a message for delivery via an application protocol layer coupled to a network subsystem to a first remote computer device, the case management system including the self-learning predictive analysis engine and the outgoing data handler interoperating directly with an http protocol layer, the case management system defining case management data associated with a plurality of cases, the defined case management data including user-configured case attributes and user-configured performance criteria, the performance criteria defining predetermined goals for associated ones of the case attributes;   storing, in a data store, configuration data including the defined case management data;   analyzing, by the self-learning predictive analysis engine, a first portion of the configuration data containing a plurality of the case attributes and a corresponding plurality of the performance criteria and associated outcomes in a first portion of the configuration data and generate a set of projected outcomes corresponding to the performance criteria;   generating, by the self-learning predictive analysis engine, correlation data which establishes for each of the case attributes a relationship with a corresponding one of the projected outcomes based on at least one of the performance criteria corresponding to respective ones of the case attribute and the projected outcome;   storing, in the data store, the correlation data;   receiving, by the self-learning predictive analysis engine, case data including one or more attributes for a corresponding specific case from the first remote computer device via the network subsystem;   identifying, by the self-learning predictive analysis engine, one or more attributes in the received case data;   generating, from the stored correlation data, a first prediction model using a first predictive algorithm and a second prediction model using a second prediction algorithm which is different than the first prediction algorithm;   testing the first and second prediction models with an actual outcome to select one prediction model that best represents the defined case management data;   identifying, by the self-learning predictive analysis engine using the selected prediction model, for the corresponding specific case, ones of the projected outcomes in the stored correlation data that are associated with the identified case attributes values in the received case data;   building, by the outgoing data handler, the message based on the identified projected outcomes for the corresponding specific case; and   providing, by the outgoing data handler, the message to the application protocol layer;   formatting the received message for delivery to the first remote computer device.   
     
     
         16 . The method of  claim 15 , further comprising:
 generating, by the case management system, the correlation data based on the first portion of the configuration data;   retrieving, by the case management system, a second portion of the configuration data which is different from the first portion of the configuration data; and   for each of one or more cases in the second portion of the configuration data: identifying, by the case management system, attributes in the case data, identifying in the stored correlation data specific projected outcomes associated with the identified case attributes, comparing the identified specific projected outcomes with respective outcomes identified for the plurality of cases in the configuration data, and verifying the specific projected outcomes with the respectively identified outcomes for the plurality of cases in the configuration data.   
     
     
         17 . The method of  claim 15 , further comprising implementing an incoming message handler which interfaces with the application protocol layer and receives the case data from the first remote computer device. 
     
     
         18 . The method of  claim 17 , further comprising implementing a command handler which is coupled to the incoming message handler and configured to route the configuration data and the case data received via the incoming message handler. 
     
     
         19 . The method of  claim 18 , further comprising:
 routing, when the command handler receives incoming configuration data, the incoming configuration data to a data manager that stores the received configuration data in the data store; and   routing, when the command handler receives incoming case data from the first remote computer device, the incoming case data to the self-learning predictive analysis engine.   
     
     
         20 . The method of  claim 15 , validating, by the self-learning predictive analysis engine, the generated correlation data against the received case data, updating the generated correlation data according to the received case data, and storing the updated correlation data.