JPWO2020201875A5

JPWO2020201875A5 -

Info

Publication number: JPWO2020201875A5
Application number: JP2021557224A
Authority: JP
Publication date: 2022-08-19

Claims

A method for accessing data records of a master data management system, said data records comprising a plurality of attributes, said method comprising:
augmenting said master data management system with one or more search engines for enabling access to said data records;
receiving a request for data at the master data management system;
identifying a set of one or more attributes of the plurality of attributes referenced in the received request;
selecting a combination of one or more of said search engines of said master data management system whose performance for searching for values of at least a portion of said set of attributes satisfies current selection rules; step to
A method comprising: processing said request using said combination of search engines; and providing at least some of the results of said processing.

receiving another request for data; updating the selection rule based on user action on the provided results, wherein the updated selection rule becomes the current selection rule; and repeating the identifying, selecting, processing, and providing steps using the current selection rule when the selection rule is selected.

said results comprising data records in said master data management system associated with each matching score obtained by a scoring engine of said search engine, said method further comprising: wherein said components include method steps, elements used to provide said results, and at least a portion of said results, said provided results being pre- 3. The method of claim 1 or 2, comprising non-duplicate data records having a weighted matching score higher than a defined score threshold.

The components involved in providing the results include the search engine, the identifying step, and the results, the method further comprising:
assigning an engine weight to each search engine of said search engines;
assigning an attribute weight to said set of attributes, said assigning step wherein said attribute weight of an attribute indicates a level of confidence with which said attribute is identified;
assigning to each data record of said results a completeness weight indicating the completeness of said data record and a freshness weight indicating freshness of said data record;
combining said respective engine weight, attribute weight, completeness weight and freshness weight for each data record of said results; and weighting said score of said data record by said combined weight. The method described in .

providing user parameters that quantify user behavior;
determining, for each component of at least some of said components, values of said user parameters and associated values of component parameters describing said component; and updating said weights assigned to said components. 5. A method according to claim 3 or 4, further comprising: using said defined association for

providing a lookup table that associates values of the user parameter with the values of the component parameters; and using the lookup table to update the weights assigned to the components. 6. The method of claim 5 , further comprising:

modeling the variation in the value of the user parameter with the value of the component parameter using a predetermined model; using the model to define update weights for the component; and using the updated weights to update the assigned weights .

a user action of the user actions includes an indication of selection of a result, the indication includes a mouse click on a displayed result of the provided results, the user parameter is the number of the clicks; A method according to any one of claims 2 to 7, comprising at least one of the frequency of said clicks and the duration of accessing a given one of said results.

For each attribute of the set of attributes, the selection rule includes:
determining, for each search engine of said search engines, a value of a performance parameter indicative of said performance of said search engine for searching for values of said attribute;
A method according to any one of claims 1 to 8 , comprising selecting said search engine having a performance parameter value higher than a predetermined performance threshold.

10. The method of claim 9 , wherein the performance parameters include at least one of the number of results and the level of matching of the results to expected values.

The selection rule uses a table that associates attributes with corresponding search engines, and the updating of the selection rule includes:
determining said value of a user parameter that quantifies said user action on the provided results of each search engine of said combination of said search engines;
using said determined value associated with each search engine of said combination of said search engines to identify said value of said user parameter that is less than a predetermined threshold; for each identified value of a parameter, defining said search engine associated with said attribute of said set of attributes and said identified value; 11. A method according to claim 9 or 10 , comprising updating said table with

A method according to any preceding claim, wherein said processing of said requests is performed in parallel by said combination of said search engines.

12. Any of claims 1-11 , wherein said combination of search engines is a ranked list of search engines, and said processing of said requests is performed sequentially according to said ranked list until a minimum number of results is exceeded. The method according to item 1.

Identifying the set of attributes includes inputting the received request into a predetermined machine learning model and receiving a classification of the request from the machine learning model, the classification being the attribute A method according to any one of claims 1 to 13 , wherein the set of

further comprising inputting said set of attributes into a predetermined machine learning model; and receiving one or more search engines that may be used to search said set of attributes from said machine learning model. , the method according to any one of claims 1 to 14 .

receiving a training set indicating a different set of one or more training attributes, each set of training attributes indicating a suitable search engine for conducting said search of said set of training attributes; 16. The method of claim 15 , further comprising the steps of receiving, labeled, and generating the machine learning model by training a predetermined machine learning algorithm using the training set. Method.

A method according to any one of the preceding claims, wherein said provided results comprise data records filtered depending on the sender of said request.

A computer program causing a processor to perform the steps of the method according to any one of claims 1-17.

A computer system for enabling access to a data record, said data record comprising a plurality of attributes, said computer system comprising a plurality of search parameters for enabling access to said data record. an engine, a user interface configured to receive a request for data, an entity identifier configured to identify a set of one or more attributes of the plurality of attributes referenced in the received request. , configured for selecting a combination of one or more of said search engines whose performance for searching for values of at least some of said set of attributes satisfies current selection rules. an engine selector, said search engine configured to process said request, said engine selector comprising a result provider configured to provide at least a portion of said results of said processing; computer system.

said results comprising data records of said computer system associated with each matching score obtained by a scoring engine of said search engine, said computer system further comprising a component responsible for providing said results; a weight provider configured to weight the matching scores according to the performance of the provided 20. The computer system of claim 19 , wherein the results obtained include non-duplicate data records having weighted matching scores higher than a predetermined score threshold.