NamSor™ is our specialized data mining software. It can predict, given a person name : likely country / region of origin, its cultural and linguistic classification, gender and spelling variants. Our innovative machine learning algorithm offers unmatched accuracy at a fine grain level, flexibility and integration capability, to
- filter through large databases and extract names,
- recognize which language or culture stands behind a name,
- accommodate all countries / regions / languages / alphabets.
For each name, in addition to classification by name origin, our unique algorithm provides several scores (strength, doubt, synthetic) that enable advanced sorting, according to specific needs. The output of our software can integrate with other data management tools, for additional data enrichment: data mining, predictive analytics, social graph analysis, semantic analysis …
Names are meaningful : we use socialinguistics to extract their semantics and deliver actionable intelligence. We proudly invented the onoma(s)tic millefeuille, a data visualization which demonstrate the importance of names, languages and culture.
As an illustration, this is the « onoma(s)tic mille-feuille » of the Olympic athletes since 1896 with in abscissa, the country -which is a known attribute- and in ordinate the name likely origin recognized by NamSor.
Some countries, such as Japan or Mongolia, look very homogeneous. Some other countries, such as the former Soviet Union, seem very heterogeneous and will stack many different onomastics classes : Russians, Ukrainians, Kazakhs, Kyrgyz, Uzbeks, Lithuanians, …
If names were meaningless, or if our software produced random figures, all we would see would be horizontal stripes.
NamSor™ helps our clients recognize names of any origin in any geography and produce meaningful insights :