Overview of the 2022 BUCC Shared Task: Bilingual Term Alignment in Comparable Specialized Corpora

Omar Adjali (1) , Emmanuel Morin (2) , Serge Sharoff (3) , Reinhard Rapp (4) , Pierre Zweigenbaum (1)

1 LISN - Laboratoire Interdisciplinaire des Sciences du Numérique
2 LS2N - Laboratoire des Sciences du Numérique de Nantes
3 University of Leeds
4 University of Mainz

The BUCC 2022 shared task addressed bilingual terminology alignment in comparable corpora. Many research groups are working on this problem using a wide variety of approaches. However, as there is no standard way to measure the performance of the systems, the published results are not comparable and the pros and cons of the various approaches are not clear. The shared task aimed at solving these problems by organizing a fair comparison of systems. This was accomplished by providing a precise definition of the task and its evaluation, and corpora and non-trivial evaluation datasets for the English-French language pair. Six runs were submitted by two teams. The obtained results are satisfactory with a top average precision of 0.28, but show that the task is not solved yet.

Keywords: Natural Language Processing; Comparable Corpora; Multilingual Processing; Term Alignment; Shared Task

Domains: Computation and Language [cs.CL]; Document and Text Processing

Omar Adjali, Emmanuel Morin, Serge Sharoff, Reinhard Rapp, Pierre Zweigenbaum. Overview of the 2022 BUCC Shared Task: Bilingual Term Alignment in Comparable Specialized Corpora. BUCC, 15th Workshop on Building and Using Comparable Corpora, Jun 2022, Marseille, France. pp.67-76. ⟨hal-03867489⟩