MadMatcher-Pro and consulting take entity matching to production, on top of an open-source core. Pro adds reliability for long-running jobs; consulting brings the team onto your data. Underneath sit three open-source tools you can also run yourself.
MadMatcher-Pro adds the reliability a long-running job needs; consulting brings the team onto your data. The fastest path from a prototype to a production matching pipeline.
The same Sparkly, Delex, and MatchFlow, plus the reliability a long-running production job needs. A multi-hour run survives a failure instead of starting from scratch.
Work directly with the team to build and tune an entity matching pipeline on your own data, from blocking strategy to a trained matcher.
Book a callThe three tools underneath, free and open source. Sparkly and Delex do blocking, MatchFlow does matching. Use one on its own or chain them into a full pipeline.
Finds candidate matches using top-k TF/IDF similarity (the BM25 variant, via Lucene). It indexes one table and searches it with the other, scaling to hundreds of millions of tuples on Spark.
Paulsen et al., VLDB 16(6), 2023. Outperforms eight state-of-the-art blockers. Read the paper ↗
Combines several blocking strategies (TF/IDF, dictionary blockers, custom rules) in one declarative program, compiled to a Spark DAG. Use it when one blocking strategy is not enough.
Trains a supervised ML matcher on labeled pairs and applies it to the blocking output. Composable functions for features, labeling, training, and prediction; runs on pandas or Spark.
Book a call to scope your matching problem with the team, or explore the tools and run the quickstart yourself.
