Internet represents one of the most important and va riegated collection of information present in the world. With the growing utilization of such a mean, the amount of the available information has growed in an exponential way but, paradoxally, the possibilities for the users to run them profitably have reduced, a phenomenon often called information overloading (V. Bush “As We May Think”, 1945). The subject of the present work is the realization of a system prototype for intelligent agents specialized in adaptive information filtering and in the research of interest information present on the Web, called Holmes, part of M&S-SW project. For such a purpose, artificial intelligence technics in the scope of the machine learning, and focus crawling technics are used. The system developed in Java with client/server architecture gives the possibility to one or more users to interact from any browser by means of a specific GUI applet. This document is subdivided as follows: in the 2nd section, we will describe the architecture and the functions of the system; subseque ntly we will make room for the user’s modeling component and for the filtering system; finally the research module of the information on internet will be illustrated and some experimental results will be proposed.
HOLMES: a Prototype for the Targeted Research of Information About Hi-Tech Companies
CAPUANO, Nicola;GAETA, Matteo;
2002
Abstract
Internet represents one of the most important and va riegated collection of information present in the world. With the growing utilization of such a mean, the amount of the available information has growed in an exponential way but, paradoxally, the possibilities for the users to run them profitably have reduced, a phenomenon often called information overloading (V. Bush “As We May Think”, 1945). The subject of the present work is the realization of a system prototype for intelligent agents specialized in adaptive information filtering and in the research of interest information present on the Web, called Holmes, part of M&S-SW project. For such a purpose, artificial intelligence technics in the scope of the machine learning, and focus crawling technics are used. The system developed in Java with client/server architecture gives the possibility to one or more users to interact from any browser by means of a specific GUI applet. This document is subdivided as follows: in the 2nd section, we will describe the architecture and the functions of the system; subseque ntly we will make room for the user’s modeling component and for the filtering system; finally the research module of the information on internet will be illustrated and some experimental results will be proposed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
