Start a new topic
Answered

Machine Learning and AI for matching

Do you use ML and AI for matching? How?


Best Answer

Semarchy xDM can leverage external AI algorithms that can participate in the matching process. Those algorithms will primarily be used to enrich or standardize the values of records before they are processed by the match rules. For example, it is possible to use face recognition APIs provided by Microsoft Azure or AWS to transform pictures on a series of attributes or hash codes that you would then use in your actual match rule. The example below demonstrates how to do this:


...
Enrich Picture (call to Azure REST API for face recognition)
... (output of enricher include hash code,  gender, etc.)
Match on AI enriched Picture:
Record1.Picture.AzureHash = Record2.Picture.AzureHash and Record1.Picture.DetectedGender = Record1.Picture.DetectedGender
... 


1 Comment

Answer

Semarchy xDM can leverage external AI algorithms that can participate in the matching process. Those algorithms will primarily be used to enrich or standardize the values of records before they are processed by the match rules. For example, it is possible to use face recognition APIs provided by Microsoft Azure or AWS to transform pictures on a series of attributes or hash codes that you would then use in your actual match rule. The example below demonstrates how to do this:


...
Enrich Picture (call to Azure REST API for face recognition)
... (output of enricher include hash code,  gender, etc.)
Match on AI enriched Picture:
Record1.Picture.AzureHash = Record2.Picture.AzureHash and Record1.Picture.DetectedGender = Record1.Picture.DetectedGender
... 


Login to post a comment