Keyword search is the most popular way to access information. In this paper we introduce a novel approach for determining the correct resources for user-supplied queries based on a hidden Markov model. In our approach the user-supplied query is modeled as the observed data and the background knowledge is used for parameter estimation. We leverage the semantic relationships between resources for computing the parameter estimations. In this approach, query segmentation and resource disambiguation are mutually tightly interwoven. First, an initial set of potential segments is obtained leveraging the underlying knowledge base; then, the final correct set of segments is determined after the most likely resource mapping was computed. While linguistic analysis (e.g. named entity, multi-word unit recognition and POS-tagging) fail in the case of keyword-based queries, we will show that our statistical approach is robust with regard to query expression variance. Our experimental results reveal very promising results.