We introduce a new logic programming paradigm -- answer set programming with uncertain facts (LPuF for short). A LPuF program is an extension of answer set programs (ASP for short). We first define the syntax of LPuF programs. Then, we propose preferred query semantics and preferred solution semantics of LPuF programs to address the issues of personal preference-dependent query generation and solution ranking in recommendation systems. Those semantics are based on the answer set semantics of logic programs. Correspondingly, we then present two algorithms which are used for computing preferred queries and preferred solutions respectively. Finally, we conclude the paper and present future works.