An Evaluation of the Concept Retrieval Annotation for Spanish-English CLEFER Parallel Corpora

This paper presents a study about the use of the concept retrieval annotation method for parallel corpora. The concept retrieval annotation method (CRA) consists of considering concepts as documents and text chunks as queries [1]. Concepts with higher similarity to text chunks are considered for generating the final semantic annotation. CRA makes use of an existing knowledge resource (KR) from which lexicons are extracted to perform the semantic annotation. Until now, CRA has been applied to mono-lingual scenarios showing a good performance over both very large collections (e.g., CALBCII-SSC) and very large lexicons (e.g., UMLS R © [2]). We have also applied this semantic annotator to different tasks in Biomedicine such as resource discovery [3], relation extraction [4], and sicentific bibliography analysis [5]. In this work, we will apply CRA in a bi-lingual scenario. For this purpose, we make use of the provided lexicons at CLEFER workshop. More specifically, we have made use of the English and Spanish lexicons. In this extended abstract, we first summarize the main features of CRM as a cross-lingual annotator, and then obtained results over the two provided parallel corpora, EMEA and MEDLINE R ©.