Automatic Keyword Extraction (AKE) is a fundamental Natural Language Processing (NLP) task that plays a critical role in various applications, including information retrieval, document summarization, and content categorization. The review’s purpose is to gather all the techniques, methodologies, and advancements available in the literature, and present them to researchers, practitioners, and developers interested in the topic, who can use this work as a starting point for their research, or to have a quick insight on what are the main trend in the automatic keyword extraction task. The review provides an in-depth analysis of traditional approaches, supervised and unsupervised, as well as emerging techniques that employ neural networks and deep learning architectures like transformers. This work also provides information regarding existing datasets and benchmarks and how to obtain them, as well as functioning code for building practical applications. At the end of the paper, there is a description of the state-of-the-art for automatic keyterm extraction, paired with a discussion on the evaluation metrics.