80 likes | 193 Views
This study explores query expansion using graph properties to improve information retrieval. Learn about transforming queries into expansion features and identifying relevant terms. Discover how this method enhances precision by up to 27% compared to traditional techniques. Topological query expansion is applied to retrieve relevant results for queries like "colored Volkswagen beetles."
E N D
Massive Query ExpansionbyExploiting Graph Properties Joan Guisado-Gámez JosepLluísLarribaPey David Domínguez Sal
Problem in Information Retrieval Query = “colored Volkswagen beetles”
Brief Introduction • Process of transforming QO into QE • Detect expansion features (terms/phrases) • What kind of expansion features? • How to obtain the expansion features?
Topological Query Expansion Query = “colored Volkswagen beetles” Context = “any model of Volkswagen beetles car of any color and year”
Expansion Features Query = “colored Volkswagen beetles” • Most important extension Features: • “Volkswagen beetle”, “Volkswagen fusca”, “VW type 1”, “Volkswagen 1200”, “Volkswagen bug”, “Volkswagen super bug”, “VW bug”,”VWKäfer” • “Volkswagen new Beetle” • “Baja bug” • “Volkswagen group”, “VW group” • H-Shifter, engine, car, automobile,….
Some Results Query = “colored Volkswagen beetles”
Conclusions • Achieves good results allowing to identify new concepts. • Precision improves up to 27% • Orthogonal to linguistic techniques. • Outperforms pseudo-relevance feedback techniques.