Genetically optimizing query expansion for retrieving activities from the web
Eleftherios Karapetsas, Yusuke Fukazawa, Jun Ota · 2012
Activities, defined as actions someone can accomplish or things one can create, are located all around the web contained in various How-To and DIY sites. In this paper we are presenting our research on a system focused on retrieving information about activities related to a simple query which is given as input. An overview of the system's design which is based on semantic query expansion is given along with detailed explanation of the optimization of the system's parameters through the use of genetic algorithms. Finally experimental results gathered from both an offline and a limited online evaluation are presented in order to test the efficiency of the system.