A three-stage learning algorithm for deep multilayer perceptron (DMLP) with effective weight initialisation based on sparse auto-encoder is proposed in this paper, which aims to overcome difficulties in training deep neural networks with limited training data in high-dimensional feature space. At the first stage, unsupervised learning is adopted using sparse auto-encoder to obtain the initial weights of the feature extraction layers of the DMLP. At the second stage, error back-propagation is used to train the DMLP by fixing the weights obtained at the first stage for its feature extraction layers. At the third stage, all the weights of the DMLP obtained at the second stage are refined by error back-propagation. Network structures an

With the rapid development of computers and network technologies, the security of information in the internet becomes compromise and many threats may affect the integrity of such information. Many researches are focused theirs works on providing solution to this threat. Machine learning and data mining are widely used in anomaly-detection schemes to decide whether or not a malicious activity is taking place on a network. In this paper a hierarchical classification for anomaly based intrusion detection system is proposed. Two levels of features selection and classification are used. In the first level, the global feature vector for detection the basic attacks (DoS, U2R, R2L and Probe) is selected. In the second level, four local feature vect

Audio classification is the process to classify different audio types according to contents. It is implemented in a large variety of real world problems, all classification applications allowed the target subjects to be viewed as a specific type of audio and hence, there is a variety in the audio types and every type has to be treatedcarefully according to its significant properties.Feature extraction is an important process for audio classification. This workintroduces several sets of features according to the type, two types of audio (datasets) were studied. Two different features sets are proposed: (i) firstorder gradient feature vector, and (ii) Local roughness feature vector, the experimentsshowed that the results are competitive to

Traffic classification is referred to as the task of categorizing traffic flows into application-aware classes such as chats, streaming, VoIP, etc. Most systems of network traffic identification are based on features. These features may be static signatures, port numbers, statistical characteristics, and so on. Current methods of data flow classification are effective, they still lack new inventive approaches to meet the needs of vital points such as real-time traffic classification, low power consumption, ), Central Processing Unit (CPU) utilization, etc. Our novel Fast Deep Packet Header Inspection (FDPHI) traffic classification proposal employs 1 Dimension Convolution Neural Network (1D-CNN) to automatically learn more representational c

The paper aims to propose Teaching Learning based Optimization (TLBO) algorithm to solve 3-D packing problem in containers. The objective which can be presented in a mathematical model is optimizing the space usage in a container. Besides the interaction effect between students and teacher, this algorithm also observes the learning process between students in the classroom which does not need any control parameters. Thus, TLBO provides the teachers phase and students phase as its main updating process to find the best solution. More precisely, to validate the algorithm effectiveness, it was implemented in three sample cases. There was small data which had 5 size-types of items with 12 units, medium data which had 10 size-types of items w

Image classification is the process of finding common features in images from various classes and applying them to categorize and label them. The main problem of the image classification process is the abundance of images, the high complexity of the data, and the shortage of labeled data, presenting the key obstacles in image classification. The cornerstone of image classification is evaluating the convolutional features retrieved from deep learning models and training them with machine learning classifiers. This study proposes a new approach of “hybrid learning” by combining deep learning with machine learning for image classification based on convolutional feature extraction using the VGG-16 deep learning model and seven class