Applying artificial intelligence in master data management has multiple technical challenges – data quality, data formats, disparate source systems, multiple entities etc. However, our innovative and scalable approach for Spark based entity resolution and fuzzy data matching has been extremely well received locally and internationally. Here are a few videos about Master Data Management, AI, Fuzzy Matching, Deduplication etc.
Agile Master Data Management At Scale Using AI
Organizations embark on master data management and spend a majority of the time integrating schemas, defining rules and algorithms and then maintaining the data pipeline at scale with growing data volumes. Traditional Master Data Management systems lack the agility to tackle the wide variety of data sources, differing schemas and matching records for millions of entities. With modern technology, we can address these challenges and supercharge the MDM journey. This webinar will address the current challenges with existing MDM systems and how AI can be leveraged to handle the variety and scale of MDM.
Customer Record Deduplication With AI – Strata Hadoop Singapore Talk
Duplicates in customer data are a painful data management issue in most organizations. Temporary workers and rule bases systems are unable to efficiently deduplicate customer records. In this talk at Strata Hadoop World, we co-presented customer name and address deduplicationalong with Mr. Dave Chan, Regional Director Business Intelligence, UBM Asia. Mr Chan talked about how the media and event industry suffers from poor customer data quality and lots of duplication in customer records. He presented the cloud based deployment of Reifier for deduplicating customer records and how they leveraged AI for effective deduplication of records with English, Chinese and Japanese characters. Mr Chan also discusses how AI based matching can be improved through human insights and expertise.
Data Mastering At Scale with AI – Strata London
What is an MDM and how does it help? What are some challenges in traditional rule based MDM systems and how can we solve them using AI and Spark? These are some of the topics in this talk about using AI to master data at scale.
Real Time Data Matching By Training From User – Spark Summit San Francisco
This talk covers our approach to learning from human stewards on how the data should be mastered. By showing few edge cases to the user, Reifier takes user input to build a machine learning model for data mastering which can be deployed at scale – batch or real time. We discuss the user experience, building training data and leveraging macine learning for master data. We also discuss the use of Elastic Search to facilitate real time lookups for record matching.
Fuzzy matching with Spark and AI – Spark Summit San Francisco
Showcasing the best of breed data science and distributed applications built over Spark, the Spark Summit covers breakthrough technologies which affect the technical landscape and demonstrate novel solutions to business problems. We talk about Reifier Fuzzy matching with Spark use cases for master data management, marketing data quality, cross selling and upselling. We also talk about how we leverage Spark for entity resolution and record linkage.