Currently available algorithms for data stream classification are all designed to handle precise data, while data with uncertainty or imperfection is quite natural and widely seen in real-life applications. Uncertainty can arise in attribute values as well as in class values. In this paper, we focus on the classification of streaming data that has different degrees of uncertainty within class values. We propose two types of ensemble based algorithms, Static Classifier Ensemble (SCE) and Dynamic Classifier Ensemble (DCE) for mining uncertain data streams. Experiments on both synthetic and real-life data set are made to compare and contrast our proposed algorithms. The experimental results reveal that DCE algorithm outperforms SCE algorithm.