Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020141968 - A SYSTEM AND METHOD FOR IMPACT ANALYSIS OF CHANGE REQUEST THAT AFFECTS DATABASE STRUCTURE THROUGH CLASSIFICIATION AND KEYWORD SIMILARITY ANALYSIS

Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

[ EN ]

CLAIMS

1. A system (100) for impact analysis of change request that affects database structure through classification and keyword similarity analysis comprising:

a client application (102) at client side for input interaction between user and server, enabling the user to configure an input interaction with the server or repository (1 10); a server side comprising:

a Requirement Relationship Knowledge Builder Engine (104) for gathering information from Requirement Book Knowledge Base, Domain Knowledge Base and Named Entity Recognition repository to generate direct tagged keyword;

a Data Mining Engine (106) for building classification, indirect tagged keyword and related word based on direct tagged keyword generated by the

Requirement Relationship Knowledge Builder Engine (104);

at least one repository (1 10) for storing named entity relationship and engine data classification, tagged keyword and related words; and

at least one Knowledge base storage (1 12) of Domain Knowledge Base and

Requirement Book Knowledge Base,

characterized in that

the server side further comprising:

an Impact Analysis Engine (108) for analysing requirements that are impacted by change request based on input from the Requirement Relationship Knowledge Builder Engine (104) and Data Mining Engine (106).

2. The system (100) according to Claim 1 , wherein the Domain Knowledge Base includes information related to keyword’s domain classification and its related words.

3. The system (100) according to Claim 1 , wherein the Requirement Book Knowledge Base includes relationship information between requirement, algorithm, database table and database column and column description; wherein each requirement comprises at least one algorithm; wherein each algorithm comprises at least one database table; and wherein each database table comprises at least one column and its description.

4. The system (100) according to Claim 1 , wherein the Domain Knowledge Base and Requirement Book Knowledge Base is used as an input to all processes and engine.

5. The system (100) according to Claim 1 , wherein the repository (1 10) includes Named Entity Recognition repository and related word repository.

6. A method (200) for impact analysis of change request that affects database structure through classification and keyword similarity analysis comprising steps of:

generating direct tagged keyword through Requirement Relationship Knowledge Builder Engine that gathers information from Requirement Book Knowledge Base, Domain Knowledge Base and repository (202); executing data mining process through Data Mining Engine by using the direct tagged keywords to build classification, indirect tagged keyword and related word (204);

storing result of classification and related word in the repository (206);

initiating impact analysis process when a change request is made (208); initiating analysis to identify related database objects requirement that would be impacted due to the change request by retrieving related requirements from the Requirement Book Knowledge Base (210);

obtaining direct tagged keywords (D{1... M}) and indirect tagged keywords (I D{1... X}) related to identified related database objects requirement by sending query to the related word repository (212);

performing impact analysis to determine highest indirect tagged keyword based on similarity analysis (214); and

generating impact analysis report based on the highest indirect tagged keyword identified, HID{1 ,... ,n} to notify indirect or other requirements that are impacted of change request (216),

characterized in that

obtaining direct tagged keywords, D{1... M} and indirect tagged keywords , ID{1...X} related to identified related database objects requirement by sending query to the related word repository (212) further comprises steps of (400):

selecting at least one indirectly tagged keyword, IDx from a list of the indirectly tagged keywords, ID{1...X} which was retrieved from the related word repository (402);

determining if the indirectly tagged keyword, IDx is higher value than value of the indirectly tagged keywords, ID{1...X} (404);

returning set of list of highest indirect keyword if the indirectly tagged keyword, IDx is higher value than the value of the indirectly tagged keywords, ID{1...X} (406);

if the indirectly tagged keyword, IDx is lower value than the value of the indirectly tagged keywords, ID{1...X}:

obtaining classification CID (1....Y) from each indirectly impacted tagged keyword, IDx (408);

retrieving related classifications, CID (1...Z) for each direct tagged keyword, Dm from the related word repository (410); retrieving indirect related words, WID[1 _ I] from the related word repository for each CIDy (412);

retrieving direct related words, WD[1....L] from the related word repository for each CDz (414);

comparing related word for each indirect WID[1 _ I], with each direct related word of WD[1....L] using cosine similarity (416); and

obtaining highest similarity average value as part of relevant indirect keyword which traces back to indirect related requirements from results obtained (418).

7. The method (200) according to Claim 6, wherein generating direct tagged keyword through Requirement Relationship Knowledge Builder Engine that gathers information from Requirement Book Knowledge Base and Named Entity Recognition repository (202) further comprises steps of (300):

generating each requirement related column by using the Requirement Book Knowledge Base (302, 302a);

tagging each Requirement Book Knowledge Base column with keywords using the Named Entity Recognition repository (304, 304a); and

generating classification and related words for each tagged keywords of each Requirement Book using Data Mining Engine (306).

8. The method (200) according to Claim 6, wherein indirect or other requirements that are impacted of change request are retrieved from Requirement Book Knowledge Base using Human Interface Device, HID which traces back to classification, tagged keyword and related word repository.

9. The method (200) according to Claim 7, wherein generating each requirement related column by using Requirement Book Knowledge Base (302, 302a) further comprises providing a set of related columns for each requirement.

0. The method (200) according to Claim 6, wherein comparing related word for each indirect WID[1 _ I], with each direct related work of WD[1....L] using cosine similarity

(416) further comprises determining similarity of words from set of related words until all related words have been selected and computed for similarity.