Event prediction using artificial intelligence

Abstract

Provided techniques manage and predict future events. For example, in a payment implementation, a supplier, at any given point in time, has multiple customer debtors that may owe payments (e.g., have outstanding invoices). Utilizing historical attributes for a given customer debtor payment predictions may be determined. By analyzing outstanding debts associated with this debtor customer an amount owed may be calculated and a predicted payment (e.g., a payment that has not yet been indicated by that debtor customer) created. Events may be provided to a second system to correlate predictions across multiple debtor collectors. Correlated information may be used to predict cash flow needs of an organization. Alternatively, optimization of help desk systems may be provided based on predictions from analysis of multiple events in an Event-driven feed back system. Provided techniques may be generalized to other applications as well.

What is claimed is: 
     
         1 . A computer-implemented method of managing and predicting payment events, the method comprising:
 determining, using a computing system, historical attributes regarding previous commitments to pay associated with a first debtor customer;   retrieving, using the computing system, one or more outstanding debts owed to a collector from the first debtor customer;   training, using the computing system, a machine learning algorithm to predict a commitment outcome by correlating the historical attributes and the commitment information;   executing, using the computing system, the machine learning algorithm to determine a predicted commitment outcome associated with at least one related outstanding debt of the one or more outstanding debts owed to the collector from the first debtor customer and generating, using the computing system and using the machine learning algorithm, the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer;   based on the predicted commitment outcome generated by the machine learning algorithm, creating and populating, using the computing system, an event associated with the at least one related outstanding debt of the first debtor customer to indicate the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer; and   providing, using the computing system, the event to an event processor system to associate the predicted commitment outcome with the at least one related outstanding debt of the first debtor customer.   
     
     
         2 . The computer implemented method of  claim 1 , further comprising:
 calculating, using the computing system, a total amount outstanding in debts relative to the first debtor customer; and   providing, using the computing system, the total amount outstanding to a collections analyst.   
     
     
         3 . The computer implemented method of  claim 1 , wherein creating the event further comprises at least one of:
 assigning, using the computing system, a first priority to the event based on a first determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a first promise is likely to be kept; or   assigning, using the computing system, a second priority to the event based on a second determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a second promise is likely to be broken.   
     
     
         4 . The computer implemented method of  claim 1 , wherein training the machine learning algorithm to predict the commitment outcome further comprises:
 correlating, using the computing system, the historical attributes, the commitment information, and a total amount associated with the related outstanding debt of the first debtor customer.   
     
     
         5 . The computer-implemented method of  claim 1 , wherein creating the event further comprises:
 creating, using the computing system, at least one of a predicted on-time payment event, a predicted delayed payment event, or a predicted very delayed payment event, wherein the predicted delayed payment event represents payment within 60 days of a payment due date and the predicted very delayed payment represents over 60 days of delay.   
     
     
         6 . The computer-implemented method of  claim 1 , wherein creating the event further comprises:
 creating, using the computing system, the event based on at least one of the at least one related outstanding debt being sent to the first debtor customer or a due date of the at least one related outstanding debt of the first debtor customer.   
     
     
         7 . The computer-implemented method of  claim 1 , further comprising:
 generating, using the computing system, a suggested action based on the event as part of the providing the event to the event processor system.   
     
     
         8 . The computer-implemented method of  claim 7 , wherein the suggested action is selected from a set of suggested actions having different severities. 
     
     
         9 . The computer-implemented method of  claim 8 , wherein the suggested action is associated with a lower severity when the at least one related outstanding debt due date is nearer a current date and a higher severity when a due date of the at least one related outstanding invoice is further from the current date, and wherein the different severities range from mild actions to general actions to strict actions with the mild actions being less severe than the general actions and the general actions being less severe than the strict actions. 
     
     
         10 . The computer-implemented method of  claim 1 , wherein the at least one related outstanding debt of the first debtor customer is at least one related outstanding invoice of the first debtor customer. 
     
     
         11 . The computer-implemented method of  claim 1 , further comprising:
 determining, using the computing system, an actual commitment outcome associated with the at least one related outstanding debt of the first debtor customer; and   retraining the machine learning algorithm based on the actual commitment outcome associated with the at least one related outstanding debt of the first debtor customer.   
     
     
         12 . A non-transitory computer readable medium comprising computer executable instructions that, when executed by one or more processing units, cause the one or more processing units to:
 determine historical attributes regarding previous commitments to pay associated with a first debtor customer;   retrieve one or more outstanding debts owed to a collector from the first debtor customer;   train a machine learning algorithm to predict a commitment outcome by correlating the historical attributes and the commitment information;   execute the machine learning algorithm to determine a predicted commitment outcome associated with at least one related outstanding debt of the one or more outstanding debts owed to the collector from the first debtor customer and generate, using the machine learning algorithm, the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer;   based on the predicted commitment outcome generated by the machine learning algorithm, create and populate an event associated with the at least one related outstanding debt of the first debtor customer to indicate the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer; and   provide the event to an event processor system to associate the predicted commitment outcome with the at least one related outstanding debt of the first debtor customer.   
     
     
         13 . The non-transitory computer readable medium of  claim 12 , wherein the instructions further comprise instructions to cause the one or more processing units to:
 calculate a total amount outstanding in debts relative to the first debtor customer; and   provide the total amount outstanding to a collections analyst.   
     
     
         14 . The non-transitory computer readable medium of  claim 12 , wherein the instructions to create the event further comprise at least one of:
 instructions to assign a first priority to the event based on a first determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a first promise is likely to be kept; or   instructions to assign a second priority to the event based on a second determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a second promise is likely to be broken.   
     
     
         15 . The non-transitory computer readable medium of  claim 12 , wherein the instructions to create the event further comprise:
 instructions to create the event based on at least one of the at least one related outstanding debt being sent to the first debtor customer or a due date of the at least one related outstanding debt of the first debtor customer.   
     
     
         16 . A computer system comprising:
 one or more processors;   a non-transitory memory communicatively coupled to the one or more processors and storing instructions executable by the one or more processors to cause the one or more processors to:
 determine historical attributes regarding previous commitments to pay associated with a first debtor customer; 
 retrieve one or more outstanding debts owed to a collector from the first debtor customer; 
 train a machine learning algorithm to predict a commitment outcome by correlating the historical attributes and the commitment information; 
 execute the machine learning algorithm to determine a predicted commitment outcome associated with at least one related outstanding debt of the one or more outstanding debts owed to the collector from the first debtor customer and generate, using the machine learning algorithm, the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer; 
 based on the predicted commitment outcome generated by the machine learning algorithm, create and populate an event to indicate the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer; and 
 provide the event to an event processor system to associate the predicted commitment outcome with the at least one related outstanding debt of the first debtor customer. 
   
     
     
         17 . The computer system of  claim 16 , wherein the instructions further comprise instructions to cause the one or more processors to:
 calculate a total amount outstanding in debts relative to the first debtor customer; and   provide the total amount outstanding to a collections analyst.   
     
     
         18 . The computer system of  claim 16 , wherein the instructions to create the event further comprise at least one of:
 instructions to assign a first priority to the event based on a first determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a first promise is likely to be kept; or   instructions to assign a second priority to the event based on a second determination that the predicted commitment outcome associated with the at least one related outstanding debt of the first debtor customer indicates a second promise is likely to be broken.   
     
     
         19 . The computer system of  claim 16 , wherein training the machine learning algorithm to predict the commitment outcome further comprises:
 instructions to correlate the historical attributes, the commitment information, and a total amount associated with the related outstanding debt of the first debtor customer.   
     
     
         20 . The computer system of  claim 16 , wherein the instructions to create the event further comprise:
 instructions to create the event based on at least one of the at least one related outstanding debt being sent to the first debtor customer or a due date of the at least one related outstanding debt of the first debtor customer.