The Main Challenge of Semi-Automatic Term Extraction Methods

Author(s):  
Merley S. Conrado ◽  
Thiago A. S. Pardo ◽  
Solange O. Rezende
2019 ◽  
Author(s):  
Antonio Šajatović ◽  
Maja Buljan ◽  
Jan Šnajder ◽  
Bojana Dalbelo Bašić

Terminology ◽  
2014 ◽  
Vol 20 (2) ◽  
pp. 151-170 ◽  
Author(s):  
Katia Peruzzo

The paper examines the possible usage of event templates derived from Frame-Based Terminology (Faber et al. 2005, 2006, 2007) as an aid to the extraction and management of legal terminology embedded in the multi-level legal system of the European Union. The method proposed here, which combines semi-automatic term extraction and a simplified event template containing six categories, is applied to an English corpus of EU texts focusing on victims of crime and their rights. Such a combination allows for the extraction of category-relevant terminological units and additional information, which can then be used for populating a terminological knowledge base organised on the basis of the same event template, but which also employs additional classification criteria to account for the multidimensionality encountered in the corpus.


Terminology ◽  
2022 ◽  
Author(s):  
Ayla Rigouts Terryn ◽  
Véronique Hoste ◽  
Els Lefever

Abstract As with many tasks in natural language processing, automatic term extraction (ATE) is increasingly approached as a machine learning problem. So far, most machine learning approaches to ATE broadly follow the traditional hybrid methodology, by first extracting a list of unique candidate terms, and classifying these candidates based on the predicted probability that they are valid terms. However, with the rise of neural networks and word embeddings, the next development in ATE might be towards sequential approaches, i.e., classifying each occurrence of each token within its original context. To test the validity of such approaches for ATE, two sequential methodologies were developed, evaluated, and compared: one feature-based conditional random fields classifier and one embedding-based recurrent neural network. An additional comparison was added with a machine learning interpretation of the traditional approach. All systems were trained and evaluated on identical data in multiple languages and domains to identify their respective strengths and weaknesses. The sequential methodologies were proven to be valid approaches to ATE, and the neural network even outperformed the more traditional approach. Interestingly, a combination of multiple approaches can outperform all of them separately, showing new ways to push the state-of-the-art in ATE.


Author(s):  
Flavius Frasincar ◽  
Wouter IJntema ◽  
Frank Goossen ◽  
Frederik Hogenboom

News items play an increasingly important role in the current business decision processes. Due to the large amount of news published every day it is difficult to find the new items of one’s interest. One solution to this problem is based on employing recommender systems. Traditionally, these recommenders use term extraction methods like TF-IDF combined with the cosine similarity measure. In this chapter, we explore semantic approaches for recommending news items by employing several semantic similarity measures. We have used existing semantic similarities as well as proposed new solutions for computing semantic similarities. Both traditional and semantic recommender approaches, some new, have been implemented in Athena, an extension of the Hermes news personalization framework. Based on the performed evaluation, we conclude that semantic recommender systems in general outperform traditional recommenders systems with respect to accuracy, precision, and recall, and that the new semantic recommenders have a better F-measure than existing semantic recommenders.


Sign in / Sign up

Export Citation Format

Share Document