Approaches, Tools, Algorithms, and Methods for Automatic Term Extraction: A Systematic Literature Mapping

Automatic Term Extraction is a key area in Natural Language Processing (NLP) focused on generating lexicographic materials essential for standardized theories and cross-lingual translations. While prior studies have surveyed the field, a systematic mapping that concurrently analyzes the relationships between approaches, tools, algorithms and methods has been lacking. This paper fills that gap by presenting a Systematic Literature Mapping (SLM) of 113 scientific papers published between 2015 and 2022. Our analysis provides a novel four-dimensional taxonomy (Approach, Domain, Evaluation, Tool Integration) for Automatic Term Extractors (ATE) and reveals a clear preference for hybrid approaches (50%), which integrate linguistic and statistical approaches, over purely statistical (38.2%) or linguistic (11.8%) ones. The findings show that Python's NLTK library is the most widely adopted tool, while algorithms like C-Value (50%) and TF-IDF (31%) dominate due to their robust performance. The unique contribution of this work is its granular, data-driven overview of the contemporary Automatic Term Extraction landscape, highlighting a critical need for more formal development methodologies and broader domain application. This study serves as a crucial resource for researchers by mapping dominant trends and identifying key research gaps.