Browsing Facultad de Matemática, Astronomía, Física y Computación by Author "Teruel, Milagro"
Now showing items 1-6 of 6
-
Combining semi-supervised and active learning to recognize minority senses in a new corpus
Cardellino, Cristian Adrián; Teruel, Milagro; Alonso i Alemany, Laura (2015)In this paper we study the impact of combining active learning with bootstrapping to grow a small annotated corpus from a different, unannotated corpus. The intuition underlying our approach is that bootstrapping includes ... -
Estudio de simplificación de oraciones con modelos actor-critic
Mazuecos Perez, Mauricio Diego (2019)La simplificación de oraciones es una tarea de Procesamiento del Lenguaje Natural que se centra en transformar escritos para que su gramática, estructura y palabras sean más sencillas de comprender, sin perder la semántica ... -
Learning slowly to learn better : curriculum learning for legal ontology population
Cardellino, Cristian; Teruel, Milagro; Alonso Alemany, Laura; Villata, Serena (2017)In this paper, we present an ontology population approach for legal ontologies. We exploit Wikipedia as a source of manually annotated examples of legal entities. We align YAGO, a Wikipedia-based ontology, and LKIF, an ... -
Legal NERC with ontologies, Wikipedia and curriculum learning
Cardellino, Cristian; Teruel, Milagro; Alonso Alemany, Laura; Villata, Serena (2017)In this paper, we present a Wikipediabased approach to develop resources for the legal domain. We establish a mapping between a legal domain ontology, LKIF (Hoekstra et al., 2007), and a Wikipediabased ontology, YAGO ... -
Minería de argumentos con aprendizaje profundo y atención
González, David Ignacio (2019)En este trabajo agregamos un mecanismo de atención a una red neuronal del estado del arte, que consiste de una red BiLSTM con embeddings de caracteres y una capa de CRF. Este modelo no sólo ha sido previamente aplicado en ... -
Reversing uncertainty sampling to improve active learning schemes
Cardellino, Cristian Adrián; Teruel, Milagro; Alonso i Alemany, Laura (2015)Active learning provides promising methods to optimize the cost of manually annotating a dataset. However, practitioners in many areas do not massively resort to such methods because they present technical difficulties and ...