Publication

Abstract : Natural language processing (NLP) and content-based image retrieval (CBIR) systems functioned efficiently with textual input. This offers a broader framework for on-going work such as text processing applications that include spam identification, content visualization, and image retrieval for the text being queried. The preprocessing of the text handling algorithm includes part of speech tagging (POS), text encoding, and text extraction function using the word embedding encoding algorithms like Word2vec, boot strapping, hidden Markov model (HMM), and so on. The derived textual input or image attribute plays a crucial role. Following this processing, the future phase can include either of the following approach, such as content-based recovery, image recovery, or speech processing. The techniques, such as content-based retrieval, image retrieval or speech processing, each of these algorithms requires their own training period in continuation with the available technical orientations. Convolutional neural network (CNN) and recurrent neural network (RNN) play a prominent role in analyzing the input dataset when it comes to testing. Generative adversarial networks (GANs) have now launched a new era in managing NLP applications in its place. This survey provides a glimpse of various algorithms that respond to text data and return with prominent accuracy involving CBIR issues and applications for text-based image synthesis. The examination of the methodologies deployed paves way for researchers to analyze and process textual data.