With the rapid development of wireless communications, the ever increasing demand for limited spectrum resources will eventually cause spectrum scarcity problem. Cognitive Radio offers a solution to this problem, which intelligently detect unused spectrum and share it, without harmful interference to primary users. It has been a research hotspot on how to analyze and utilize licensed spectrum to meet the increasing need for bands. Aiming at such a problem, this paper presents a CR system model and then proposes the scheme for spectrum data mining from the large-scale of TV spectrum sensing data. We designed three levels of information mining: statistical information, space clustering, and sequential pattern mining. The simulation results illustrate the effectiveness of the data mining scheme.