ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-09082015-112958


Tipo di tesi
Tesi di laurea magistrale
Autore
ALFEO, ANTONIO LUCA
URN
etd-09082015-112958
Titolo
A biologically-inspired approach to assess patent-based indicators trends via adaptive marker-based stigmergy
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
COMPUTER ENGINEERING
Relatori
relatore Ing. Cimino, Mario Giovanni Cosimo Antonio
relatore Prof.ssa Martini, Antonella
relatore Ing. Appio, Francesco Paolo
relatore Prof.ssa Vaglini, Gigliola
Parole chiave
  • Smart Specialization
  • Regional Innovation
  • Patent-Based Indicators
  • Parametric Adaptation
  • Marker-Based Stigmergy
  • Differential Evolution
  • Trend Analysis
Data inizio appello
25/09/2015
Consultabilità
Completa
Riassunto
Regional innovation is more and more considered an important enabler of welfare. It is no coincidence that the European Commission has started looking at regional peculiarities and dynamics, in order to focus Research and Innovation Strategies for Smart Specialization towards effective investment policies. In such context, this work aims to support policy makers in the analysis of innovation-relevant trends.
For this purpose, a software system is designed and developed to assess innovation trends indicators. In contrast with conventional knowledge-based design, here the approach is biologically-inspired and characterized by self-organization of information, robustness and flexibility of the solution.
To determine the dynamics of a set of technological innovation indicators about specialization-diversificationhe the exploitation of the European patent application database is provided.
After dataset pre-processing, the resulting time series are converted into spatiotemporal-aggregated behavioral tracks, via marker-based stigmergy, in order to enable knowledge to self-organize and emerge. This allows a marking structure appearing and staying spontaneously at runtime, when some local dynamism occurs. At a second level of processing, similarity evaluation is performed between prototype derived from different timed tracks in order to assess behavior deviations. The purpose of this approach is to overcome an explicit modeling behaviors that is very inefficient to be managed in such varying scenario.
The effectiveness of the proposed system has been tested, validated and experimented on real-world scenarios.
File