Here are some details about Zingg. Hope you find them useful
◦Training Data: Zingg uses a findTrainingData phase to search for edge cases in the data, which can then be labeled by the user to train the model. This approach minimizes user effort and allows models to be built and deployed quickly.
◦ User Feedback: The cycle of findTrainingData and labeling is designed to be iterative, allowing Zingg models to continuously improve through user feedback. This iterative approach enables the models to become smarter and more accurate over time
◦ Zingg works on AWS Glue: https://aws.amazon.com/blogs/big-data/entity-resolution-and-fuzzy-matches-in-aws-glue-using-the-zingg-open-source-library/
◦ Data Sources and Sinks: Zingg supports various data sources and sinks, including Databricks, Snowflake, JDBC (Postgres, MySQL), AWS S3, Cassandra, MongoDB, Neo4j, Parquet, BigQuery, and Exasol
◦ Cloud Support: Zingg can be run on cloud platforms like AWS, Azure, Databricks, and Fabric
◦ Scalability: Zingg can process large sample sizes and by judiciously selecting training data, Zingg can achieve good performance even with large overall sample sizes, as it focuses on the most informative examples
