Most advertisement systems widely used in Internet try to improve advertisement process by targeting specific groups of potential customers. Many systems exploit the information directly provided by the user and the data collected by monitoring user activities in order to built accurate user profiles, which determines the success of the advertisement process.This paper presents a solution to the problem of targeting advertisement information when minimal knowledge about anonymous internet user is given. In particulary as, for example, in the case of search engines, the user remains anonymous and his interaction with the service can be very limited. In this case the information about him is sparse and based only on the keywords and the data submitted by the HTTP request. The proposed architecture is based on the use of predefined profiles and the computation of fuzzy similarities in order to match the observed user with appropriate target profiles. The notion of fuzzy similarity presented here is based on the theoretical framework of the Łukasiewicz structure, which guarantees the correctness of the approach.
Fuzzy Matching of User Profiles for a Banner Engine
MILANI, Alfredo;
2004-01-01
Abstract
Most advertisement systems widely used in Internet try to improve advertisement process by targeting specific groups of potential customers. Many systems exploit the information directly provided by the user and the data collected by monitoring user activities in order to built accurate user profiles, which determines the success of the advertisement process.This paper presents a solution to the problem of targeting advertisement information when minimal knowledge about anonymous internet user is given. In particulary as, for example, in the case of search engines, the user remains anonymous and his interaction with the service can be very limited. In this case the information about him is sparse and based only on the keywords and the data submitted by the HTTP request. The proposed architecture is based on the use of predefined profiles and the computation of fuzzy similarities in order to match the observed user with appropriate target profiles. The notion of fuzzy similarity presented here is based on the theoretical framework of the Łukasiewicz structure, which guarantees the correctness of the approach.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
