Deep neural based name entity recognizer and classifier for English language

Abstract

Named entity recognition (NER) is an important and very effective for the Machine Translation, Retrieval (IR), Information Extraction (IE) from huge corpus, Question Answering (QA) system, text Mining and text clustering and etc. NER help us to classify or identify the Noun and its types such place /location, people, department, Ministry, organization, times and etc. The huge data available on social Media, websites, news channels and many more sources can be classified so that it can be used in research for NLP processes such as in Machine Translation, Speech Technology, Information Extraction and etc. To process this huge data or corpus we propose recent techniques of Machine Learning and Deep Neural Network. The Deep Neural Network approach will help to identify the Named entity (NE) from huge corpus or text by training the corpus using Word2vec approach. On the basis of fetched tokens and tag. We categorize these tokens into different Grammar categories based of cosine similarity concept of Deep Neural Network. Cosine similarity help to find the tag of unknown token or phases by finding its neared Vectors which are not trained earlier in Word2vec database. We have used the supervised learning (SL) techniques to train the network.