Comparison of NER Performance Using Word Embedding
Miran Seok, Hye-Jeong Song, Chan-Young Park, Jong-Dae Kim, Yu-seop Kim · Advanced science and technology letters · 2015
Recent studies in NER use the supervised machine learning. This study used CRF as a learning algorithm, and applied word embedding to feature for NER training. Word embedding is helpful in many learning algorithms of NLP, indicating that words in a sentence are mapped by a real vector in a lowdimension space. As a result of comparing the performance of multiple techniques for word embedding to NER, it was found that CCA (85.96%) in Test A and Word2Vec (80.72%) in Test B exhibited the best performance.