Learning to rank for web image retrieval based on genetic programming

Ranking is a crucial task in information retrieval systems. This paper proposes a novel ranking model named WIRank, which employs a layered genetic programming architecture to automatically generate an effective ranking function, by combining various types of evidences in Web image retrieval, including text information, image-based features and link structure analysis. This paper also introduces a new significant feature to represent images: Temporal information, which is rarely utilized in the current information retrieval systems and applications. The experimental results show that the proposed algorithms are capable of learning effective ranking functions for Web image retrieval. Significant improvement in relevancy obtained, in comparison to some other well-known ranking techniques, in terms of MAP, NDCG@n and D@n.