Abstract

NLG has been used in a variety of areas such as report generation, machine translation, information generation, and word summarization. These problems require the creation of long essays from short essays, which are very interesting and difficult. This requires strong thinking and reasoning skills that match human performance. The generation of macroeconomic news and short-term news was done through deep neural network techniques. Automatic text generation is important in natural language processing and artificial intelligence. Articles can be made to write stories, weather reports, and even poems. It is also important for machine translation, text summarization, query answering, and information systems. NLP relies on cognitive representation studies to understand how people communicate using patterns in language and text. The main goal is to create algorithms that can translate human language between machines and humans. In this article, the latest developments in document publishing and automated publishing from 2009 to 2022 were reviewed. As a result, the characteristics of the research area are said to have evolved significantly over time. Common datasets, methods, and capabilities of integration are discussed. Several research areas from various types of research were proposed to inspire future research in thematic areas, especially for essay writing. It is good to develop an effective NLP solution for producing essays, articles, publications, and future content. In particular, the most important applications are publishing articles, publishing political articles, formulating and marking test questions, robots, and conference reports.