System, method, and computer program for identifying significant attributes of records

• US 10,067,990 B1
• Filed: 03/03/2016
• Issued: 09/04/2018
• Est. Priority Date: 03/03/2016
• Status: Active Grant

• Current Assignee: Amdocs Development Limited (Amdocs Ltd.)
• Original Assignee: Amdocs Development Limited (Amdocs Ltd.)
• Inventors: Boothman, Catherine, Cogan, Peter John, Ferguson, Paul Andrew, Rutowski, Tomasz
• Primary Examiner(s): Stevens, Robert
• Application Number:
  US15/060,520
• Time in Patent Office: 915 Days
• Field of Search: 700722
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/046 Forward inferencing; Production systems
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

A system, method, and computer program product are provided for identifying significant attributes of records (e.g. journeys, etc.). A plurality of records are stored, including a plurality of events with a plurality of attributes. Further, the attributes of the events are processed, utilizing machine learning. To this end, at least one of the attributes are identified as being significant, based on the processing. Such identified at least one attribute may then be displayed, etc.

Claims

1. A computer program product embodied on a non-transitory computer readable medium, comprising computer code for:

storing, in a database, a plurality of records, each of the records including information relating to a sequence of events occurring over time and involving interactions between a customer and on-line and off-line systems, and each of the records having a plurality of attributes relating to the events and including;

an intent,

a duration,

a customer type, and

and an outcome resulting from the sequence of events, the outcome selected from a plurality of outcomes that include a success and a failure;

analyzing the records to identify a pattern in the sequences of events that more often than not lead to a certain outcome of the plurality of outcomes;

processing, utilizing at least one processor, the attributes of the records having the identified pattern, utilizing machine learning, to;

identify which attributes of the records lead to the certain outcome of the plurality of outcomes, and

identify values of the identified attributes that lead to the certain outcome of the plurality of outcomes; and

adjusting at least one business rule to enable a subsequent different sequence of events for at least one customer, based on the identified values of the identified attributes that lead to the certain outcome of the plurality of outcomes.

2. The computer program product of claim 1, wherein the success involves payment of a bill and the failure involves a churn event.

3. A method, comprising:

storing, in a database, a plurality of records, each of the records including information relating to a sequence of events occurring over time and involving interactions between a customer and on-line and off-line systems, and each of the records having a plurality of attributes relating to the events and including;

an intent,

a duration,

a customer type, and

and an outcome resulting from the sequence of events, the outcome selected from a plurality of outcomes that include a success and a failure;

analyzing the records to identify a pattern in the sequences of events that more often than not lead to a certain outcome of the plurality of outcomes;

processing, utilizing at least one processor, the attributes of the records having the identified pattern, utilizing machine learning, to;

identify which attributes of the records lead to the certain outcome of the plurality of outcomes, and

identify values of the identified attributes that lead to the certain outcome of the plurality of outcomes; and

adjusting at least one business rule to enable a subsequent different sequence of events for at least one customer, based on the identified values of the identified attributes that lead to the certain outcome of the plurality of outcomes.

4. A system, comprising:

one or more hardware processors for;

storing, in a database, a plurality of records, each of the records including information relating to a sequence of events occurring over time and involving interactions between a customer and on-line and off-line systems, and each of the records having a plurality of attributes relating to the events and including;

an intent,

a duration,

a customer type, and

and an outcome resulting from the sequence of events, the outcome selected from a plurality of outcomes that include a success and a failure;

analyzing the records to identify a pattern in the sequences of events that more often than not lead to a certain outcome of the plurality of outcomes;

processing the attributes of the records having the identified pattern, utilizing machine learning, to;

identify which attributes of the records lead to the certain outcome of the plurality of outcomes, and

identify values of the identified attributes that lead to the certain outcome of the plurality of outcomes; and

adjusting at least one business rule to enable a subsequent different sequence of events for at least one customer, based on the identified values of the identified attributes that lead to the certain outcome of the plurality of outcomes.