Query Expansion Using Word Embedding, Ontology and Natural Language Processing

Query Expansion (QE) is the art of reconstructing specific queries to expand validation presentation, especially in the data mining process in a requirement understanding environment. Expanding requirements is one of the techniques involved in finding information. In the search engine environment, the query extension includes the evaluation of the value of the construction and the extension of search queries to match new documents. In natural language processing (NLP), word embedding is a term used in textbook parsing, usually as a real-valued vector that encodes the meaning of adjacent words in the vector. It is assumed that the space will be analogous in meaning. Word embedding can be achieved using a set of language models and point literacy methods where vocabulary words or expressions are mapped to vectors of real numbers. For query expansion, one method used is natural language processing through word embedding. Other approaches are ontology, machine learning, and deep learning for automatic query expansion. This paper proposes a hybrid approach for query expansion by combining NLP and ontology through word embedding.