The papers in this special issue are aimed at giving some hints on which classes of problems — based on structured data — may appropriately be faced via machine learning tools able to process graphs. The main appeal of graphical approaches is, in fact, that of devising an automatic way for processing data, based upon the morphological interrelationships (or interconnections) present within the data. Actually, classical pattern recognition systems have proven to be effective for data which contain an inherent, identifiable organization such as image data (which is organized by location within a visual rendering) and time series data (which is organized by time), whereas structural pattern recognition overcomes this limitation, and guarantees that learning involves the whole information (numerical, categorical, topological) attached to the data. The usefulness of structural pattern recognition systems, however, is limited as a consequence of fundamental complications associated with the implementation of machine learning models that deal with such complex data. Those models are both interesting by themselves, since the study of their theoretical properties constitute the boundary of the present research in connectionism, and for the advanced real-world applications which they allow to cope with.
Bianchini, M., Scarselli, F. (a cura di). (2009). Neurocomputing - Special Issue on Pattern Recognition in Graphical Domains. Elsevier.
Neurocomputing - Special Issue on Pattern Recognition in Graphical Domains
BIANCHINI, MONICA;SCARSELLI, FRANCO
2009-01-01
Abstract
The papers in this special issue are aimed at giving some hints on which classes of problems — based on structured data — may appropriately be faced via machine learning tools able to process graphs. The main appeal of graphical approaches is, in fact, that of devising an automatic way for processing data, based upon the morphological interrelationships (or interconnections) present within the data. Actually, classical pattern recognition systems have proven to be effective for data which contain an inherent, identifiable organization such as image data (which is organized by location within a visual rendering) and time series data (which is organized by time), whereas structural pattern recognition overcomes this limitation, and guarantees that learning involves the whole information (numerical, categorical, topological) attached to the data. The usefulness of structural pattern recognition systems, however, is limited as a consequence of fundamental complications associated with the implementation of machine learning models that deal with such complex data. Those models are both interesting by themselves, since the study of their theoretical properties constitute the boundary of the present research in connectionism, and for the advanced real-world applications which they allow to cope with.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/34768
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo