We extend Latent Dirichlet Allocation (LDA) by introducing a topic model, Hierarchically Dual Latent Dirichlet Allocation (HDLDA), for contexts in which one type of documents (e.g., search queries) are semantically related to another type of documents (e.g., search results). In the context of...

We develop a flexible content-based search model that links the content preferences of search engine users to query search volume and click-through rates. Content preferences are defined over latent topics that describe the content of search queries and search result descriptions. Moreover, our...