Enterprise management system using artificial intelligence and machine learning for technology analysis and integration

Patent metadata

Field	Value
Publication number	US-11973657-B2
Application number	US-202017076412-A
Country	US
Kind code	B2
Filing date	Oct 21, 2020
Priority date	Oct 21, 2020
Publication date	Apr 30, 2024
Grant date	Apr 30, 2024

Abstract

A system may receive enterprise information associated with a client enterprise. The system may select, using an industry analysis model, a set of queries associated with obtaining status information that is associated with a technology profile of the client enterprise. The system may generate client data that is associated with the enterprise information and the status information. The system may convert, using a matrix factorization technique, the client data associated with the client enterprise to a client matrix. The system may convert, using the matrix factorization technique, reference data associated with reference enterprises to a reference matrix. The system may determine, based on a comparison of the client matrix and the reference matrix, a set of scores associated with technology metrics of the technology profile. The system may perform an action associated with the client enterprise based on the set of scores.

First claim

What is claimed is:

1. A method, comprising:
obtaining, by a device, an industry analysis model that is configured to analyze enterprises associated with a particular industry;
receiving, by the device, enterprise information associated with a client enterprise;
selecting, by the device and using the industry analysis model, a set of queries associated with obtaining status information that is associated with a technology profile of the client enterprise;
providing, by the device, the set of queries to a client device to obtain the status information;
receiving, by the device and from the client device, the status information;
generating, by the device, client data that is associated with the enterprise information and the status information;
converting, by the device and using a matrix factorization technique, the client data associated with the client enterprise to a client matrix;
selecting, by the device and based on the client data, reference data associated with reference technology profiles of reference enterprises that are associated with the particular industry;
converting, by the device and using the matrix factorization technique, the reference data to a reference matrix;
determining, by the device and based on a comparison of the client matrix and the reference matrix, a set of scores associated with technology metrics of the technology profile; and
performing, by the device, an action associated with the client enterprise based on the set of scores.

2. The method of claim 1 , wherein the industry analysis model is a machine learning model that is trained to identify technology metrics associated with a particular industry based on historical data associated with a plurality of enterprises associated with the particular industry.

3. The method of claim 2 , wherein the machine learning model is configured to utilize a natural language processing technique to obtain the historical data from at least one of:
historical surveys of one or more of the plurality of enterprises that include one or more of the set of queries;
historical articles associated with one or more of the plurality of enterprises that describe types of technologies associated with the technology profile; or
historical research documents associated with one or more of the plurality of enterprises that describe the types of technologies associated with the technology profile.

4. The method of claim 1 , wherein selecting the reference data comprises:
determining similarity scores between the enterprise information and reference enterprise information associated with the reference enterprises;
determining that the similarity scores satisfy a threshold in a comparison of the reference enterprises and characteristics of the client enterprise;
identifying, in a reference data structure, entries in the reference data structure that are associated with the reference enterprises; and
selecting the reference data from the entries.

5. The method of claim 4 , wherein the similarity scores are determined using a K-nearest neighbor classification analysis.

6. The method of claim 1 , wherein the technology metrics correspond to levels of usage of particular types of technology by the client enterprise, and
wherein the set of scores are determined based on the levels of usage of the particular types of technology based on corresponding reference levels of usage of the particular types of technology that are identified in the reference technology profiles.

7. The method of claim 1 , further comprising:
prior to performing the action, identifying a score of the set of scores that satisfies a threshold,
wherein the score of the set of scores is associated with a level of usage of a particular technology; and

determining, based on the score of the set of scores satisfying the threshold, that the level of usage of the particular technology should be increased,
wherein the action is performed to cause the client enterprise to automatically increase the level of usage of the particular technology.

8. The method of claim 1 , wherein performing the action comprises at least one:
generating a recommendation associated with a particular technology based on the set of scores;
providing the recommendation to a management device associated with the client enterprise;
obtaining an offer associated with a transaction involving usage of a particular technology based on the set of scores;
providing the offer to the management device to facilitate the transaction;
causing a usage transaction associated with a particular technology to increase or decrease usage of the particular technology; or
storing the enterprise information, the status information, and the recommendation in a reference data structure associated with the industry analysis model.

9. A device, comprising:
one or more memories; and
one or more processors, communicatively coupled to the one or more memories, configured to:
receive enterprise information associated with a client enterprise;
determine, from the enterprise information, enterprise characteristics associated with the client enterprise;
select, based on the enterprise characteristics, a set of queries associated with obtaining status information that is associated with a technology profile of the client enterprise;
provide the set of queries to a client device to obtain the status information;
receive, from the client device, the status information;
generate client data that is associated with the enterprise characteristics and technology metrics identified in the status information;
select, based on the client data, reference data associated with reference technology profiles of reference enterprises that are associated with a particular industry;
generate, from the client data and using a matrix factorization technique, a client matrix that is based on the technology metrics and the enterprise characteristics;
generate, from the reference data and using the matrix factorization technique, a reference matrix that is based on corresponding technology metrics and corresponding enterprise characteristics of the reference enterprises;
determine, based on a comparison of the client matrix and the reference matrix, a recommendation associated with usage of a particular technology of the technology profile; and
perform an action associated with the recommendation and the client enterprise.

10. The device of claim 9 , wherein the set of queries are selected using a machine learning model that is trained to identify sets of technology metrics associated with one or more key performance indicators of the particular industry based on historical data associated with a plurality of enterprises associated with the particular industry.

11. The device of claim 9 , wherein the one or more processors, when selecting the reference data, are configured to:
determine similarity scores between the enterprise characteristics and reference enterprise characteristics associated with the reference enterprises;
determine that the similarity scores satisfy a threshold in a comparison of characteristics of the reference enterprises and characteristics of the client enterprise;
identify, in a reference data structure, entries in the reference data structure that are associated with the reference enterprises; and
select the reference data from the entries.

12. The device of claim 11 , wherein the enterprise characteristics and reference enterprise characteristics are associated with at least one of:
a financial metric;
a location;
an industry;
a market metric;
a growth metric; or
an enterprise size metric.

13. The device of claim 9 , wh

Assignees

• Accenture Global Solutions Ltd

Inventors

• Prasad Rajendra Tanniru
• Ghosh Bhaskar
• Kulkarni Aditi
• Vijayaraghavan Koushik M
• JAGANNATHAN Purnima
• Jagtap Parul
• Jayaram Sangeetha
• Parameswar Badrinath
• Mishra Manas
• Dsouza Jeffson Felix
• Goenka Gaurav
• Sood Gaurav
• Senapati Pradeep
• Ramaswamy Vaijayanthi
• Tharayil Ranjith
• Zurale Mahesh
• Venkataraman Ramanathan

Classifications

• H04L41/14 Primary

  Network analysis or design · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• G06F18/24147

  Distances to closest patterns, e.g. nearest neighbour classification · CPC title

• G06N5/04

  Inference or reasoning models · CPC title