With the extensive application of wireless sensor networks (WSNs), it has become increasingly vulnerable of several types of attacks. DoS attack is one of the most dangerous attack that threaten wireless sensor network security and could have a serious impact in WSNs functionalities. Recently, Intrusion detection systems is one of the crucial security components. We propose in this work, an approach based on machine and Deep learning for DoS attack detection. We have evaluated and analyzed the efficiency of three learning models separately, including Deep Neural Network (DNN), Random Forest and Decision Tree, using the standard metrics of evaluation such as accuracy, precision, F1-score and recall. An approach Ensemble learning based majority voting was introduced taking as input, the outputs result of the three distinct Machine Learning (ML) classifiers in order to enhance the level of security. Our model was carried out using a well-known dataset WSN-DS, intended for wireless sensor networks, containing four types of Dos attacks: Blackhole, Grayhole, Flooding and TDMA. The experiment result demonstrate the effectiveness of our approach in DoS detection with high accuracy achieved close to perfect. Our approach result demonstrate that using the ensemble learning based majority-voting technique works better than each ML architecture independently.