International Journal of Advanced Innovative Technology in Engineering (IJAITE)

---

Overview of Word Embeddings in NLP: A Survey

Mr. Sameer Yadavrao Thakur, Dr. K. H. Walse, Dr. V. M. Thakare

Abstract :

This is a survey paper on “Word embedding in NLP”. For this, we reviewed 21 research papers in which WE were used as the main technique. The Paper has seven different sections. Word Embedding in NLP is gaining overwhelming responses nowadays because of its important properties for representing words. Representing words, phrases, sentences, and documents for various machine learning activities is a crucial task of NLP. Therefore, it is necessary to convert or represent textual inputs to key in for the machine for discovering knowledge about the text which is available around various systems. There are various studies in which we can employ word embedding techniques in common NLP activities like extraction and retrieval of information, analysis of sentient or opinion, sentence classification, question-answer retrieval, identifying the syntax and semantics of words concerning context and sense, relation modeling, learning using privileged information, retrieving information from bilingual, cross-lingual, multilingual, detection of the erroneous word, misspelled, out of vocabulary word, the semantics of words from the web and social media and many other applications. This survey paper gives details regarding word embedding methods and approaches used to address various recent issues and challenges of NLP tasks.

Keywords :

Full Text :

Download PDF

DOI :

Cite this paper :

References :

---