Scientometric Analysis and Implementation of Word Embedding Techniques in Natural Language Processing

Understanding a language's syntactic, semantic, and expressive components is the goal of natural language processing, also known as NLP. This field of study involves processing a language to comprehend its components. The developments in natural language processing (NLP) have benefited various industries, such as natural machine translation, named entity recognition, sentiment analysis, and chatbots, to name just a few examples. The process of natural language processing (NLP) may be broken down into two primary components: the first is the transformation of the input text (raw data) into a numerical format (vectors or matrices), and the second is the building of models for the processing of numerical data. Both aspects are necessary for understanding the topic. The primary focus of this work is investigating how the discipline of natural language processing (NLP) has evolved from rule-based, statistical representations to learning representations that are more sensitive to context. This study aims to provide an overview of the approaches that allow text representations, with a particular emphasis on embeddings for text of varying lengths and, more significantly, on works that go beyond word embeddings simultaneously. The analysis of individual words is typically more accessible than the analysis of entire text sections because several extra elements come into play. Due to this reason, representations of lengthier pieces of text can be created through various methodologies, which leverage more excellent information compared to what is done for single words. In addition to providing a report on the most recent research trend in this field, the purpose of this study is to provide an overview of the development of word embedding methodology and specific descriptions of typical methods. With the help of this paper, one can acquire a concise guide to comprehending the fundamental concept of word embedding and its progression.