loading page

Master Data Management using Machine Learning Techniques: MDM Bot
  • Tapan Behera ,
  • BS Panda
Tapan Behera

Corresponding Author:[email protected]

Author Profile

Abstract

In today’s world of loosely coupled and distributed applications, communication and data centralization play a vital role. In large organizations, there are many applications and many teams that work on the data of customers, products, suppliers, and other business entities. Each application and team maintain a separate copy of the data, which means any application requiring this information needs to contact several other systems to obtain the most recent data for that entity. Therefore, this would increase the number of multiple lookups on various systems, resulting in an increase in the throughput of the application. To avoid this, the organization needs a centralized system called ”Master Data Management” that will maintain all the golden data across all the systems within the organization. Any system that is updated with new information will also be sent to MDM through events, and MDM will then be responsible for communicating with and updating the sub-systems within the organization’s Eco-System. Event-Driven Architecture will be used to communicate between distributed applications to achieve and implement Master Data Management. In recent years, there have been a lot of issues discovered in the MDM System between the data sync-up of all the sub-systems, resulting in a tedious process of identifying manually all the failures of the events, reprocessing the event  information, correcting the data,and re-synchronizing the other sub-systems with all the previous ones. To avoid this manual process and to address the issues mentioned above, we introduce the ”MDM Bot” in this article. Artificial Intelligence and Machine Learning capability are included in the proposed MDM Bot.It is an excellent value-added Bot for an Enterprise System that helps and saves a lot of time and money.