Pasar al contenido principal

APRENDIZAJE AUTOMATICO PARA ANALISIS BIG DATA DE FUENTES DOCUMENTALES HETEROGENEAS

TEC2017-83838-R

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos I+D+i Retos de Investigación
Año convocatoria 2017
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD CARLOS III DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Optimized Update/Prediction Assignment for Lifting Transforms on Graphs

  • Martínez Enríquez, Eduardo
  • Cid Sueiro, Jesús
  • Díaz de María, Fernando
  • Ortega Gómez, Román Antonio
Transformations on graphs can provide compact representations of signals with many applications in denoising, feature extraction or compression. In particular, lifting transforms have the advantage of being critically sampled and invertible by construction, but the efficiency of the transform depends on the choice of a good bipartition of the graph into update (U) and prediction (P) nodes. This is the update/prediction (U=P) assignment problem, which is the focus of this paper. We analyze this problem theoretically and derive an optimal U=P assignment under assumptions about signal model and filters. Furthermore, we prove that the best U=P partition is related to the correlation between nodes on the graph and is not the one that minimizes the number of conflicts (connections between nodes of same label) or maximizes the weight of the cut. We also provide experimental results in randomly generated graph signals and real data from image and video signals that validate our theoretical conclusions, demonstrating improved performance over state of the art solutions for this problem., This work was supported in part by NSF under Grant CCF-1018977 and in part by the Spanish Ministry of Economy and Competitiveness under Grants TEC2014-53390-P, TEC2014-52289-R, TEC2016-81900-REDT/AEI and TEC2017-83838-R, Publicado
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Corpus Viewer: NLP and ML-based Platform for Public Policy Making and Implementation, Corpus Viewer: una plataforma basada en PLN y Aprendizaje Automático para diseño e implementación de política pública

  • Pérez-Fernández, David
  • Arenas-García, Jerónimo
  • Samy, Doaa
  • Padilla-Soler, Antonio
  • Gómez-Verdejo, Vanesa
Corpus Viewer is a production service developed by the State Secretary for Digital Advancement (SEAD) within the framework of the National Language Technologies Plan (Plan TL), promoted by the same State Secretary. Corpus Viewer relies on Natural Language Processing (NLP), Machine Learning (ML) and Machine Translation (MT) to analyze structured metadata and unstructured textual data in large document corpora. The platform allows the decision maker and the policy implementer the possibility of analyze R&D&i information space (mainly patents, scientific publications and public aids) for evidence and knowledge-based policy making and implementation. In this paper, we describe the main functionalities of the platform and enumerate the techniques it is based on, which include a variety of methods like document topic modeling and graph analysis.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Landscaping Language Technologies using Topic Modeling and Graph Analysis: Overview of the Spanish Contribution, Caracterización del sector de Tecnologías del Lenguaje mediante modelado de tópicos y análisis de grafos: Visión general de la participación española

  • Samy, Doaa
  • Pérez-Fernández, David
  • Arenas-García, Jerónimo
This paper aims at landscaping the Human Language Technologies (HLT) sector by applying topic modeling and graph analysis to study the scientific literature in ACL Anthology with special emphasis on the Spanish participation. The analysis takes into account the structured and unstructured data to offer an overview of the HLT landscape in Spain identifying main underlying themes and its evolution in the last years compared to the international HLT community. Results obtained are represented through an interactive visualization to allow the exploration of the HLT landscape in the time frame 1983-2018.