Browse Prior Art Database

System and method to ensure historical database quality with cloud-based service for data alalysis Disclosure Number: IPCOM000247691D
Publication Date: 2016-Sep-27
Document File: 2 page(s) / 150K

Publishing Venue

The Prior Art Database


We present in this article how to ensure database quality with engine or cloud-based service for data analysis/consistency using external/internal inputs

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 2

System and method to ensure historical database quality with cloud-based service for data alalysis

Description of the environment the problem arose:
Database quality needs to be enforced anywhere and anytime,

As an example, assume we get access to a new database with population by country and Nigeria shows 90 M.

    Total of all countries is OK. We actually go on with an error as real population is 180 M, but we had no clue to find out the mistake.

    The day before it was correct with 180 M (but these numbers have been replaced), but all men have disappeared in the new daily load.

    We see in this example that from a global point of view the value is correct but in details is not coherent anymore.

    Description of the problem to be solved in this context and constraints associated with: When creating a database, the administrator can specify field attributes like "string" or "number", can set up lower and/or upper limits or use active database

    However, overtime, numbers are moving and can break limits (and alert the administrator every time) as these limits are neither changed nor auto-adaptive

    Best way to solve the issue is to ignore the alerts.
=> we need to guarantee database quality with up-to-date rules adjusted according the context using:

    - A method and system that gets probabilistic functions according to domain (from an external cognitive system).

    It allows any database to self-check that new data are aligned to existing data using above probabilistic rules.

    - A method and system that checks consistency between series that can be compared (using internal system based on correlation, history and similitude) (see example).

    - A method and system that detects abnormal data and allows correction (if appropriate).

    - A method and system that defines new rules that the administrator can accept/reject.

The following dimensions are taken into account :

    Definition of statistical laws based on a time function with...