Abstract:

The field of Information Retrieval has produced some of the most impactful Computer Science research: from the early beginning of exploring how to search articles and books in libraries, to the current focus on searching the World Wide Web. In this talk, I will make the case for Ad Retrieval as a new distinct sub-discipline of Information Retrieval focusing on retrieving online advertisements. Online advertising affects virtually every web user and has grown into a $20 billon industry. As with the Web corpus, the structure of the online ads is substantially different than any other previously studied corpus. The queries used in selecting online ads can also substantially differ from the commonly explored short textual queries, as for example when selecting advertisements for a given web page or a context of a user. These differences require reexamination of many conclusions of traditional IR, as document analysis, query expansion, scoring and length normalization, and performance evaluation. The talk will outline some of the main challenges of Ad Retrieval and discuss how to engage the SIGIR community in the exploration of this new frontier of Information Retrieval.