This paper proposes a data mining based sleep wave and sleep stage classification scheme. Sleep wave classification have a greater use in diagnosis of health conditions. Classification of sleep wave and sleep stage may be put into use in clinics where there is a lack of specialist physicians. Various data mining techniques used in this work are Artificial Neural Network (ANN), Support Vector Machines (SVM), Adaptive neuro-fuzzy inference system (ANFIS), k-nearest neighbor (kNN), naive Bayes classifier (NBC), Linear/Quadratic discriminate analysis (LDA/QDA) and Decision tree (DT). The dataset was obtained from physionet sleep EDF dataset repository. The data set includes horizontal EOG, submental-EMG and EEG signals which are inputs to the data mining based network. Output of data mining based network composed of different classes corresponds to a certain sleep wave like alpha, beta, delta and theta and sleep stages like Awake (Stage 0), Stage I, II, III, IV and REM sleep. Accuracy of the classification is 99% for sleep wave and sleep stage. Since DT has the highest accuracy and it is easy to use, it is chosen for sleep wave classification.