Detect person names in text: Part 2 (Technical)

Jan ProcházkaDeep Learning, Personal Data, Whitepaper

Neural network architecture

In Detect Person Names in Text: Part 1 (Results), we benchmarked our new named entity recognizer (NER) against popular open source alternatives, such as Stanford NER, Stanza and SpaCy. Today we dig a little deeper into the NER architecture and technical details. First, recall our main NER objectives. In short, we require our NER to be practical, rather than just … Read More

Detect person names in text: Part 1 (Results)

Jan ProcházkaDeep Learning, Personal Data, Whitepaper

F1 scores for different software doing personal name detection.

Detecting people’s names is part and parcel of PII discovery. Traditional techniques like regexps and keywords don’t work, because the set of all names is too varied. How do open source Named Entity Recognition (NER) engines compare, and can we do better? This Part 1 has NER results and benchmarks. There’s also Part 2 with technical neural network details. Developing … Read More