KAS-term: Extracting Slovene Terms from Doctoral Theses via Supervised Machine Learning
This addresses term extraction for Slovene academic texts, but it is incremental as it builds on existing methods with specific improvements.
The paper tackled term extraction from Slovene academic texts by creating a dataset and using supervised machine learning, achieving an AUC of 0.736 for multi-word terms compared to 0.590 with the best single statistic.
This paper presents a dataset and supervised learning experiments for term extraction from Slovene academic texts. Term candidates in the dataset were extracted via morphosyntactic patterns and annotated for their termness by four annotators. Experiments on the dataset show that most co-occurrence statistics, applied after morphosyntactic patterns and a frequency threshold, perform close to random and that the results can be significantly improved by combining, with supervised machine learning, all the seven statistic measures included in the dataset. On multi-word terms the model using all statistics obtains an AUC of 0.736 while the best single statistic produces only AUC 0.590. Among many additional candidate features, only adding multi-word morphosyntactic pattern information and length of the single-word term candidates achieves further improvements of the results.