In this paper, we present a system designed to discover recommendation fuzzy rules useful to provide personalized link suggestions to the users of a Web site. The system is mainly based on two processes. A fuzzy clustering process is applied to identify user categories by grouping users with similar interests. Then, a neuro-fuzzy strategy is applied to derive a set of recommendation fuzzy rules. A tool for the proposed system provides a wizard-based interface made of a sequence of panels that support users in the overall rule extraction process. An illustrative example is provided to show the performance of the system through the use of the developed tool.
A system for deriving a neurofuzzy recommendation model
CASTELLANO, GIOVANNA;FANELLI, Anna Maria;
2009-01-01
Abstract
In this paper, we present a system designed to discover recommendation fuzzy rules useful to provide personalized link suggestions to the users of a Web site. The system is mainly based on two processes. A fuzzy clustering process is applied to identify user categories by grouping users with similar interests. Then, a neuro-fuzzy strategy is applied to derive a set of recommendation fuzzy rules. A tool for the proposed system provides a wizard-based interface made of a sequence of panels that support users in the overall rule extraction process. An illustrative example is provided to show the performance of the system through the use of the developed tool.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.