System and method for prioritization of support requests

Patent metadata

Field	Value
Publication number	US-11574016-B2
Application number	US-202016778013-A
Country	US
Kind code	B2
Filing date	Jan 31, 2020
Priority date	Jan 31, 2020
Publication date	Feb 7, 2023
Grant date	Feb 7, 2023

Abstract

Methods, information handling systems and computer readable media are disclosed for determining a priority score for a pending support request document. According to one embodiment, a method includes receiving current support request information from within a pending support request document and accessing current additional information associated with the pending support request document. The method further includes associating a set of parameter values with the pending support request document, wherein the values within the set of parameter values are based on information within one or both of the current support request information or the current additional information. The method continues with determining a priority score corresponding to the set of parameter values, where determining the priority score comprises applying a machine learning model developed using previous support request information and previous additional information associated with previously-resolved support request documents, and assigning the priority score to the pending support request document.

First claim

What is claimed is:

1. A method, comprising
receiving current support request information, where the current support request information is contained within a pending support request document;
accessing current additional information associated with the pending support request document;
associating a set of parameter values with the pending support request document, wherein the parameter values in the set of parameter values are based on information within one or both of the current support request information or the current additional information, wherein associating the set of parameter values comprises applying natural language processing to a text string within the pending support request document, and applying a machine learning classification model to a result of the natural language processing;
training the machine learning classification model using previous support request information and previous additional information associated with previously-resolved support request documents;
assigning sentiment values for the pending support request document, by a sentiment analysis module using the machine learning classification model, wherein the set of parameter values comprises the sentiment values, and wherein the set of parameter values comprises a sentiment value for the pending support request document;
determining a priority score corresponding to the set of parameter values, wherein determining the priority score comprises applying a machine learning model developed using the previous support request information and the previous additional information associated with the previously-resolved support request documents; and
assigning the priority score to the pending support request document.

2. The method of claim 1 , further comprising:
comparing the priority score to additional priority scores assigned to respective additional pending support request documents;
based on a result of the comparing, establishing a priority sequence for response to the pending support request document and the additional pending support request documents; and
sending the pending support request document and the additional pending support request documents to one or more client information handling systems for response, according to the priority sequence.

3. The method of claim 1 , wherein
the set of parameter values comprises a hybrid parameter value; and
associating the set of parameter values comprises accessing results of a clustering algorithm to obtain the hybrid parameter value.

4. The method of claim 1 , wherein the machine learning model comprises a neural network.

5. The method of claim 1 , wherein the current additional information comprises information retrieved from one or more data stores during an update of the machine learning model.

6. An information handling system, comprising:
one or more processors;
one or more non-transitory computer-readable storage media coupled to the one or more processors; and
a plurality of instructions, encoded in the one or more computer-readable storage media and configured to cause the one or more processors to
receive current support request information, where the current support request information is contained within a pending support request document;
access current additional information associated with the pending support request document;
associate a set of parameter values with the pending support request document, wherein the parameter values in the set of parameter values are based on information within one or both of the current support request information or the current additional information, wherein associating the set of parameter values comprises applying natural language processing to a text string within the pending support request document, and applying a machine learning classification model to a result of the natural language processing;
train the machine learning classification model using previous support request information and previous additional information associated with previously-resolved support request documents;
assign sentiment values for the pending support request document, by a sentiment analysis module using the machine learning classification model, wherein the set of parameter values comprises the sentiment values, and wherein the set of parameter values comprises a sentiment value for the pending support request document;
determine a priority score corresponding to the set of parameter values, including applying a machine learning model developed using the previous support request information and the previous additional information associated with the previously-resolved support request documents; and
assign the priority score to the support request document.

7. The information handling system of claim 6 , wherein the plurality of instructions is further configured to cause the one or more processors to:
compare the priority score to additional priority scores assigned to respective additional pending support request documents;
based on a result of the comparing, establish a priority sequence for response to the pending support request document and the additional pending support request documents; and
send the pending support request document and the additional pending support request documents to one or more client information handling systems for response, according to the priority sequence.

8. The information handling system of claim 6 , wherein
the set of parameter values comprises a hybrid parameter value; and
the plurality of instructions is further configured to cause the one or more processors to access results of a clustering algorithm to obtain the hybrid parameter value, as a part of associating the set of parameter values with the pending support request document.

9. The information handling system of claim 6 , wherein the machine learning model comprises a neural network.

10. The information handling system of claim 6 , wherein the current additional information comprises information retrieved from one or more data stores during an update of the machine learning model.

11. A non-transitory computer readable storage medium having program instructions encoded therein, wherein the program instructions are executable to:
receive current support request information, where the current support request information is contained within a pending support request document;
access current additional information associated with the pending support request document;
associate a set of parameter values with the pending support request document, wherein the parameter values in the set of parameter values are based on information within one or both of the current support request information or the current additional information, wherein associating the set of parameter values comprises applying natural language processing to a text string within the pending support request document, and applying a machine learning classification model to a result of the natural language processing;
train the machine learning classification model using previous support request information and previous additional information associated with previously-resolved support request documents;
assign sentiment values for the pending support request document, by a sentiment analysis module using the machine learning classification model, wherein the set of parameter values comprises the sentiment values, and wherein the set of parameter values comprises a sentiment value for the pending support request document;
determine a priority score corresponding to the set of parameter values, including applying a machine learning model developed using the previous support request information and the previous additional informa

Assignees

• Dell Products Lp

Inventors

• Bikumala Sathish Kumar
• Lena Marcio Fragoso Stumpf
• Nagarajegowda Deepak

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06F16/90332 Primary

  Natural language query formulation or dialogue systems · CPC title

• H04L41/16 Primary

  using machine learning or artificial intelligence · CPC title

• G06N20/00

  Machine learning · CPC title