Issue extraction based on ticket mining

Patent metadata

Field	Value
Publication number	US-2016110723-A1
Application number	US-201514918244-A
Country	US
Kind code	A1
Filing date	Oct 20, 2015
Priority date	Oct 20, 2014
Publication date	Apr 21, 2016
Grant date	—

Abstract

Described herein are a method and a system for extracting issues based on ticket mining. In one implementation, a plurality of tickets comprising descriptions of the issues in computing systems are received. The received descriptions are then cleaned by removing unwanted details. Upon cleaning, the clean descriptions are mapped with descriptions stored in service catalog data to obtain unmapped clean descriptions. In an example, the unmapped clean descriptions include one of user-generated descriptions, system-generated descriptions, and both the user-generated descriptions and the system-generated descriptions. For the user-generated descriptions; the issues are extracted by pre-processing the user-generated descriptions, determining keywords from the processed unmapped clean descriptions, constructing n-grams of keywords from the extracted keywords, and extracting the n-grams of keywords as the issues present in the computing systems.

First claim

I/We claim:

1 . A method for extracting issues based on ticket mining, the method comprising:
receiving a plurality of tickets comprising descriptions of the issues in computing systems; cleaning the descriptions by removing unwanted details from the descriptions of the plurality of tickets; mapping the clean descriptions with descriptions stored in service catalog data retrieved from a database, to obtain unmapped clean descriptions, wherein the unmapped clean descriptions include one of user-generated descriptions, system-generated descriptions, and both the user-generated descriptions and the system-generated descriptions; and extracting the issues from the unmapped clean descriptions, wherein for the user-generated descriptions, the extracting comprises:
pre-processing the unmapped clean descriptions;
determining keywords from the processed unmapped clean descriptions;
constructing n-grams of keywords from extracted keywords, the n-grams of keywords representing issues described in the extracted keywords; and
extracting the n-grams of keywords as the issues present in the computing systems.

2 . The method as claimed in claim 1 , wherein for the system-generated descriptions, the extracting comprises:
clustering the unmapped clean descriptions into separate clusters based on similarity; labeling each of the clusters with a label that represents the unmapped clean descriptions of a cluster, extracting the labels as the issues present in the computing systems.

3 . The method as claimed in claim 1 , wherein the cleaning comprising retaining domain-specific words using knowledge of a plurality of domains.

4 . The method as claimed in claim 1 , wherein the service catalog data comprises descriptions associated with a set of issues for which resolution steps are known.

5 . The method as claimed in claim 1 , wherein the preprocessing comprises one of stemming, synonym detecting, and spelling correcting of the unmapped clean descriptions.

6 . The method as claimed in claim 1 , wherein the constructing comprises:
grouping similar keywords based on similarity between the keywords; and labeling each group using the n-grams of keywords.

7 . An issue extraction system for extracting issues based on ticket mining, the system comprising:
a processor; a data preparation module, coupled to the processor, to:
receive a plurality of tickets comprising descriptions of the issues in computing systems; and
clean the descriptions by removing unwanted details from the descriptions of the plurality of tickets;
a mapping module, coupled to the processor, to map the clean descriptions with descriptions present in service catalog data retrieved from a database, for obtaining unmapped clean descriptions, wherein the unmapped clean descriptions include one of user-generated descriptions, system-generated descriptions, and both the user-generated descriptions and the system-generated descriptions; and an issue extraction module coupled to the processor, wherein for the user-generated descriptions, the issue extraction module is adapted to:
pre-process the unmapped clean descriptions,
determine keywords from the processed unmapped clean descriptions,
construct n-grams of keywords from extracted keywords, the n-grams of keywords representing issues described in the extracted keywords, and
extract the n-grams of keywords as the issues present in the computing systems.

8 . The system as claimed in claim 7 , wherein the issue extraction module comprises a clustering module, and wherein for the system-generated descriptions, the clustering module is adapted to:
cluster the unmapped clean descriptions into separate clusters based on similarity; label each of the clusters with a label that represents the unmapped clean descriptions of a cluster; and extract the labels as the issues present in the computing systems.

9 . The system as claimed in claim 7 , wherein the data preparation module cleans the descriptions by retaining domain-specific words using knowledge of a plurality of domains.

10 . The system as claimed in claim 7 , wherein the service catalog data comprises descriptions associated with a set of issues for which resolution steps are known.

11 . The system as claimed in claim 7 , wherein the issue extraction module comprises a pre-processing module for pre-processing the unmapped clean descriptions by one of stemming, synonym detecting, and spelling correcting process.

12 . The system as claimed in claim 7 , wherein the issue extraction module comprises a n-gram construction module to construct the n-grams of keywords from the extracted keywords by:
grouping similar keywords based on the similarity between the keywords; and labeling each group using the n-grams of keywords.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Shimpi Vikrant Vikas
• Natu Maitreya
• Sadaphal Vaishali
• Kulkarni Vaishali

Classifications

• G06Q10/063

  Operations research, analysis or management · CPC title

• G06Q30/016 Primary

  After-sales · CPC title

• G06F16/38

  Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually · CPC title

• G06F17/30722

  Physics · mapped topic