Using Stanford NER and Illinois NER to Detect Malay Named Entity Recognition

Abstract

The goal of NER is to detect named entities in an open document. Many techniques are used to solve the NER problem. Most Malay Named Entity Recognition uses rule based and gazette to tag the names for each entity. In this paper, we tested online news articles using Stanford NER and Illinois NER to measure the capability of these NER tools to detect Malay Named Entities. The results are computed using the CoNLL evaluation metric. Stanford NER tends to produce higher results on F1 and Precision compared to Illinois NER. In the future, more NER systems will be evaluated to measure the compatibility of the tools to recognize Malay Named Entities.