Traffic classification technique is an essential tool for network and system security in the complex environments such as cloud computing based environment. The state-of-the-art traffic classification methods aim to take the advantages of flow statistical features and machine learning techniques, however the classification performance is severely affected by limited supervised information and unknown applications. In unsupervised methods, different group of similar items called clusters are generated, but these clusters are need to be identified. For this we need some additional supervised information. In Classifier, a pre-labeled set of training instances are used to train the classifier. To make an accurate classifier this set of pre-labeled instances must be large, but it is impossible since new applications are emerging day by day. Also supervised method never detects unknown flows or intrusions. To tackle these problems, I designed a novel semi-supervised approach that integrates the advantages of both supervised and semi-supervised methods. This technique is applied over the real-time data to simulate the proper behavior of this new methodology.