System And Method For Recommending Automation Solutions For Technology Infrastructure Issues

Patent metadata

Field	Value
Publication number	US-2020162312-A1
Application number	US-201816237469-A
Country	US
Kind code	A1
Filing date	Dec 31, 2018
Priority date	Nov 19, 2018
Publication date	May 21, 2020
Grant date	—

Abstract

A system and method to intelligently formulate automation strategies for technology infrastructure operations are disclosed. The system and method include analyzing infrastructure issue data from support tickets and predicting automation solutions. A cost-benefit analysis is then performed on the automation solutions. Solutions can be ranked and recommended according to the cost-benefit analysis.

First claim

We claim:

1 . A method of predicting automation solutions for technology infrastructure issues, the method comprising:
analyzing information related to a plurality of technology infrastructure issues; extracting a set of data features corresponding to the plurality of technology infrastructure issues; identifying a first set of technology infrastructure issues and predicting a first automation solution for the first set of technology infrastructure issues; identifying a second set of technology infrastructure issues and predicting a second automation solution for the second set of technology infrastructure issues; estimating a first cost-savings amount associated with implementing the first automation solution for the first set of technology infrastructure issues; estimating a second cost-savings amount associated with implementing the second automation solution for the second set of technology infrastructure issues; and ranking the first automation solution higher than the second automation solution when the first cost-savings amount is greater than the second cost-savings amount.

2 . The method according to claim 1 , wherein the method includes the steps of receiving a plurality of technology infrastructure support tickets; and
retrieving the information related to the plurality of technology infrastructure issues from the plurality of technology infrastructure support tickets.

3 . The method according to claim 1 , wherein analyzing information related to the plurality of infrastructure issues includes identify sequences of related technology infrastructure issues.

4 . The method according to claim 1 , wherein extracting the set of data features includes:
converting the information related to the plurality of technology infrastructure issues into vectors in a vectors space; and calculating the cosine similarity between the vectors in the vector space and a unit vector in the vector space.

5 . The method according to claim 1 , wherein predicting the first automation solution includes using a Multinomial Naïve Bayes analysis and a Gradient Boosting Machine.

6 . The method according to claim 1 , wherein the method includes using information from an infrastructure knowledge base, the infrastructure knowledge base including information about a set of activities and automation solutions for the set of activities.

7 . The method according to claim 6 , wherein the method further includes periodically updating the infrastructure knowledge base.

8 . The method according to claim 1 , wherein the method includes producing a plurality of solutions corresponding to different groups of technology infrastructure issues, ranking the plurality of solutions, and recommending a subset of ranked solutions.

9 . The method according to claim 1 , wherein the method includes predicting a technology associated with the first automation solution, and wherein the technology is predicted by analyzing verbs in the information related to the plurality of technology infrastructure issues.

10 . The method according to claim 9 , wherein the technology is a software scripting technology.

11 . The method according to claim 1 , wherein extracting features includes applying a term frequency-inverse document frequency matrix to a set of words associated with one or more technology infrastructure issues.

12 . A non-transitory computer-readable medium storing software comprising instructions that are executable by one or more device processors to predict automation solutions for technology infrastructure issues by:
analyzing information related to a plurality of technology infrastructure issues; extracting a set of data features corresponding to the plurality of technology infrastructure issues; identifying a first set of technology infrastructure issues and predicting a first automation solution for the first set of technology infrastructure issues; identifying a second set of technology infrastructure issues and predicting a second automation solution for the second set of technology infrastructure issues; estimating a first cost-savings amount associated with implementing the first automation solution for the first set of technology infrastructure issues; estimating a second cost-savings amount associated with implementing the second automation solution for the second set of technology infrastructure issues; and ranking the first automation solution higher than the second automation solution when the first cost-savings amount is greater than the second cost-savings amount.

13 . The non-transitory computer-readable medium of claim 12 , wherein analyzing the information includes determining sequences of infrastructure issues.

14 . The non-transitory computer-readable medium of claim 13 , wherein determining sequences includes using a random forest algorithm.

15 . The non-transitory computer-readable medium of claim 13 , wherein determining sequences includes using an unsupervised rule induction and affinity analysis.

16 . The non-transitory computer-readable medium of claim 12 , wherein extracting the set of data features includes using a non-linear classification algorithm.

17 . The non-transitory computer-readable medium of claim 12 , wherein the instructions are executable to predict automation solutions by producing a plurality of solutions corresponding to different groups of technology infrastructure issues, ranking the plurality of solutions, and recommending a subset of ranked solutions.

18 . A system for predicting automation solutions for technology infrastructure issues, the system comprising:
a device processor; and a non-transitory computer readable medium storing instructions that are executable by the device processor to:
analyze information related to a plurality of technology infrastructure issues;
extract a set of data features corresponding to the plurality of technology infrastructure issues;
identify a first set of technology infrastructure issues and predict a first automation solution for the first set of technology infrastructure issues;
identify a second set of technology infrastructure issues and predict a second automation solution for the second set of technology infrastructure issues;
estimate a first cost-savings amount associated with implementing the first automation solution for the first set of technology infrastructure issues;
estimate a second cost-savings amount associated with implementing the second automation solution for the second set of technology infrastructure issues; and
rank the first automation solution higher than the second automation solution when the first cost-savings amount is greater than the second cost-savings amount.

19 . The system according to claim 18 , wherein to estimate a first cost-savings amount, the instructions are executable by the device processor to calculate the manual effort spent performing an activity associated with the first automation solution.

20 . The system according to claim 19 , wherein the instructions are executable by the device processor to predict a break-even time for recovering costs associated with the first automation solution.

Assignees

• Accenture Global Solutions Ltd

Inventors

• Balasubramanian Ramkumar
• Bhutani Sandeep
• Pilla Chandrasekhar
• GUPTA Shallu
• Maddula Sekhar Naga Venkata

Classifications

• H04L41/069

  using logs of notifications; Post-processing of notifications · CPC title

• H04L41/5074

  Handling of user complaints or trouble tickets · CPC title

• H04L41/0686

  Additional information in the notification, e.g. enhancement of specific meta-data · CPC title

• G06F40/216

  using statistical methods · CPC title

• G06F16/24578

  using ranking · CPC title