Skip to main content

Resultados relacionados

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

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Aggregation-based information retrieval system for geospatial data catalogs

  • Lacasta, Javier
  • Lopez-Pellicer, F. Javier
  • Espejo-García, Borja
  • Nogueras-Iso, Javier
  • Zarazaga-Soria, F. Javier
Geospatial data catalogs enable users to discover and access geographical information. Prevailing solutions are document oriented and fragment the spatial continuum of the geospatial data into independent and disconnected resources described through metadata. Due to this, the complete answer for a query may be scattered across multiple resources, making its discovery and access more difficult. This paper proposes an improved information retrieval process for geospatial data catalogs that aggregates the search results by identifying the implicit spatial/thematic relations between the metadata records of the resources. These aggregations are constructed in such a way that they match better the user query than each resource individually.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Agricultural recommendation system for crop protection

  • Lacasta Miguel, Javier
  • Lopez-Pellicer, F. Javier
  • Espejo-García, Borja
  • Nogueras-Iso, Javier
  • Zarazaga-Soria, F Javier
Pests in crops produce important economic loses all around the world. To deal with them without damaging people or the environment, governments have established strict legislation and norms describing the products and procedures of use. However, since these norms frequently change to reflect scientific and technological advances, it is needed to perform a frequent review of affected norms in order to update pest related information systems. This is not an easy task because they are usually human-oriented, so intensive manual labour is required. To facilitate the use of this information, this work proposes the construction of a recommendation system that facilitates the identification of pests and the selection of suitable treatments. The core of this system is an ontology that models the interactions between crops, pests and treatments.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

End-to-end sequence labeling via deep learning for automatic extraction of agricultural regulations

  • Espejo-Garcia, Borja
  • Lopez-Pellicer, Francisco J.
  • Lacasta, Javier
  • Piedrafita Moreno, Ramón
  • Zarazaga-Soria, F. Javier
In the European Union, production standards in the form of legal regulations play an important role in farming. Because of the increasing amount of regulations, it is desirable to transform human-oriented regulations into a set of computer-oriented rules to provide decision support through the Farm Management Information System. To obtain the logical structure of rules, automatically labeling their meaningful information is necessary. In this work, we evaluate the performance of 8 different state-of-the-art deep learning architectures to develop an end-to-end sequence labeler for phytosanitary regulations. This sequence labeler extracts different meaningful information items to determine which pesticides can be applied to a crop, the place of the treatment, when it can be applied, and the maximum number of applications. The architectures evaluated do not require feature engineering and, hence, they are applicable to the agricultural regulations of different countries. The best system is a neural network that uses character embeddings, Bidirectional Long short-term memory and Softmax. It achieves a performance of 88.3% F 1 score.