Upgrade Your Fuzzy Name Matching
Rosette is trusted by the world’s largest and most critical banking and security systems.
Percentage of correct matches testing against a dataset with 7,571 person names, with at least 10 variants for each name.
✓ Enhance your current system
✓ Machine learning algorithms
✓ Scalable
✓ Explainable
✓ Flexible deployment options
Select Customers
See Rosette in action
Learn how our algorithm works and the underlying capabilities that allow for a readable and accurate score.
Accurate
- Reduce false positives — When a name is erroneously matched to a watch list record, time and money is wasted in manually following up on each hit
- Reduce false negatives — When a match is missed, your screening system has failed its mission and consequences can be very serious
- Simultaneously considers 13 name variations including cross-lingual matching in 21 languages — including Arabic, Cyrillic, and Asian scripts
- Employs a hybrid of methods including phonetics, rules, deep learning, and statistical models
- Special algorithms for more accurate Chinese and Japanese organization name matching.
Fast
- Slow systems get ignored, ensure your organization does a thorough check of every name by deploying Rosette
- Consistent response time for long or short names; names with two components or more than five
- User configurable to increase speed or recall to meet the needs of your task or data.
Active development
- At least six software releases per year
- Fast response to support requests.
Flexible
- Deploys as a plugin to Elasticsearch, Solr, or via cloud or on-premise (SDK or server)
- User configurable and tunable to dataset characteristics
- Confidence score produced for each match, ability to weight any number of other identity attribute fields.
“…answering, ‘Is that the same person or not based on a name?’ is subjective, and the amount of time for a human can vary greatly, and accuracy varies greatly. That’s where being able to apply sophisticated software makes a huge difference.”