Overview of CAPITEL Shared Tasks at IberLEF 2020: Named Entity Recognition and Universal Dependencies Parsing

TL;DR

This paper reports on the CAPITEL-EVAL shared task at IberLEF 2020, focusing on Spanish named entity recognition and dependency parsing using a new annotated newswire corpus, involving seven teams and 13 submissions.

Contribution

It introduces a new annotated Spanish newswire corpus and evaluates multiple systems on NER and dependency parsing tasks in a shared evaluation setting.

Findings

Multiple systems participated, demonstrating diverse approaches.

The shared task provided benchmark results for Spanish NER and parsing.

The new corpus facilitated evaluation of NLP tools for Spanish.

Abstract

We present the results of the CAPITEL-EVAL shared task, held in the context of the IberLEF 2020 competition series. CAPITEL-EVAL consisted on two subtasks: (1) Named Entity Recognition and Classification and (2) Universal Dependency parsing. For both, the source data was a newly annotated corpus, CAPITEL, a collection of Spanish articles in the newswire domain. A total of seven teams participated in CAPITEL-EVAL, with a total of 13 runs submitted across all subtasks. Data, results and further information about this task can be found at sites.google.com/view/capitel2020.