There have been several methods developed to construct ensembles. Some of these methods, such as Bagging and Boosting are meta-learners, i.e. they can be applied to any base classifier. The combination of methods should be selected in order that classifiers cover each other weaknesses. In ensemble, the output of several classifiers is used only when they disagree on some inputs. The degree of disagreement is called diversity of the ensemble. Another factor that plays a significant role in performing an ensemble is accuracy of the basic classifiers. It can be said that all the procedures of constructing ensembles seek to achieve a balance between these two parameters, and successful methods can reach a better balance. The diversity of the members of an ensemble is known as an important factor in determining its generalization error. In this paper, we present a new approach for generating ensembles. The proposed approach uses Bagging and Boosting as the generators of base classifiers. Subsequently, the classifiers are partitioned by means of a clustering algorithm. We introduce a selection phase for construction the final ensemble and three different selection methods are proposed for applying in this phase. In the first proposed selection method, a classifier is selected randomly from each cluster. The second method selects the most accurate classifier from each cluster and the third one selects the nearest classifier to the center of each cluster to construct the final ensemble. The results of the experiments on well-known datasets demonstrate the strength of our proposed approach, especially applying the selection of the most accurate classifiers from clusters and employing Bagging generator.

This paper focuses on how students access web pages in a task specific information retrieval. An investigation on how students search the web for their current needs was carried out and students’ behavioural characteristics as they surf the internet to answer some given online multiple choice questions was collected. Twenty three students participated in the study and a number of behavioural characteristics were captured. Camtasia studio 7 was used to record their searching activity. The result shows that 328 web pages were visited by the students, and among the parameters captured, the time spent on the search task has a stronger correlation with the students’ performance than any other captured parameter. The time spent on a document can be used as a good implicit indicator to infer learner’s interest in a context based recommender system.

Traditional distributed database transaction applications within large organizations often involve a large number of resources. In this case, people and DDBMSs distributed over a wide geographic area, may introduce conflict between heterogeneous systems. Web services (WS) provide solution for this problem since WS have an independent platform, independent language, and independent object model. This work presents WS application to access heterogeneous and distributed database via horizontal data fragments that is designed to be reliable, flexible and scalable. It describes the setup of SOAP server and applications based on the SOAP for end user client. In addition, it allows the publishing of WS descriptions to submit user requests (goal) to retrieve the required information. Here we evaluate the functional, behavior and performance of WS among possible different alternatives with real-time and execution parameters. Implementation details and case study experiments are presented along with the corresponding results.

The wireless communication nature of remotely deployed sensor nodes make the attacks more easily to be happened in wireless sensor networks (WSNs). But traditional security algorithms are infeasible in WSNs due to the limited computing, communication power, storage, band width and energy of sensor nodes. So energy efficient secure data aggregation schemes are necessary in resource constrained WSNs. Concealed Data Aggregation (CDA) based on privacy homomorphism (PH) gives a critical solution for energy efficient secure data aggregation in WSNs. PH based algorithms allow aggregation to be happened on cipher texts. Thus, it eliminates the power consuming decryption operations at the aggregator node for the data aggregation and further encryption for the secure transmission of aggregated data. It also avoids the aggregator node from the burden of keeping the secret key information and thereby it achieves energy efficiency and reduces the frequency of node compromise attacks in aggregator nodes. Among the CDA techniques, asymmetric PH based CDA techniques are exploring due to their combination with elliptic curve cryptography having reduced key size. We present an overview of asymmetric concealed data aggregation techniques that achieve both end to end data confidentiality and non delayed data aggregation.

Generally, wireless systems suffer from Carrier Frequency Offsets (CFO), which results either from the receiver’s oscillator’s impairments or Doppler shifts. The frequency offset between transmitter and receiver local oscillator is a main drawback of OFDM systems. A frequency drift in the receiver’s oscillator would result in a frequency offset in the received signal, this offset consists of two parts: a Fine offset, as a fraction of OFDM subcarrier spacing, and a Coarse offset that is equal to an integer multiple of the subcarrier spacing. Coarse offsets results in two problems; firstly, the modulation symbols are located on the wrong frequency bins, secondly, the pilots used for channel estimation are misplaced leading to a faulty channel estimate and severe signal distortion by the Equalizer. Fine Offsets result in a loss of orthogonality and thus ICI (Inter-Carrier Interference), as the frequency bin of a certain modulation symbol would contain some information about the neighboring symbol bin.

Automatic person identification is an important task in computer vision and related applications. Multimodal biometrics involves more than two modalities. The proposed work is an implementation of person identification fusing face and ear biometric modalities. We have used PCA based neural network classifier for feature extraction from the images. These features are fused and used for identification. PCA method was found better if the modalities were combined. Identification was made using Eigen faces, Eigen ears and their features. These were tested over own created database.

Regarding to the influence of robots in the various fields of life, the issue of trusting to them is important, especially when a robot deals with people directly. One of the possible ways to get this confidence is adding a moral dimension to the robots. Therefore, we present a new architecture in order to build moral agents that learn from demonstrations. This agent is based on Beauchamp and Childress’s principles of biomedical ethics (a type of deontological theory) and uses decision tree algorithm to abstract relationships between ethical principles and morality of actions. We apply this architecture to build an agent that provides guidance to health care workers faced with ethical dilemmas. Our results show that the agent is able to learn ethic well.

The design patterns are the reusable component used in the development of the software, which delivers enhanced quality software to the end users. The design patterns are available for user interface, mobile applications, text classification and so on. There are no design patterns for internet banking applications. This motivated to mine the design patterns for internet banking application from the document of Business Process Management (BPM) by using the qualitative research technique. The nonfunctional quality attribute of software architecture is enhanced by using the design patterns. In this paper the mined two patterns are presented namely substitute pattern and communication pattern for internet banking application.

Self Organizing Wireless Sensor Networks (WSN) is an emergent and challenging technology that is applicable to various real life scenarios. Different routing protocols in the WSN have been proposed over the years. In this type of network the major concern is the energy constraint sensor nodes that operate on limited battery power. Hence energy efficient routing algorithm in WSN need to be developed in order to address the battery power constraint of the sensor nodes. Minimizing the communication overhead during the data transmission and reception can considerably reduce the energy requirement of the sensor nodes. Redundant Binary Number System (RBNS) is one of the energy efficient techniques that can reduce the communication overhead by reducing the number of 1’s required to communicate in a data packet. In our proposed work the RBNS communication technique is applied with an existing popular routing protocol in WSN to achieve an energy efficient routing protocol. The algorithm has been successfully implemented in a simulated environment and the result that has obtained demonstrates the significant enhancement of network lifetime of the sensor network.

Fuzzy logic controller (FLC) is an important nonlinear controller in an uncertain dynamic system’s parameters. This controller is used to control of nonlinear dynamic systems particularly for spherical motor, because it has a suitable control performance and it is a stable. Conversely pure fuzzy logic controller is a high-quality intelligent nonlinear controller; it has two important problems; reliability and robustness in uncertain dynamic parameter. To increase the reliability and robustness, this research is focused on applied feedback linearization method in pure fuzzy logic controller. In this research the nonlinear equivalent dynamic (equivalent part) formulation problem in uncertain condition is also solved by combine pure fuzzy logic control and feedback linearization method. In this method feedback linearization theorem is applied to fuzzy logic controller to increase the stability, reliability and robustness, which it is based on nonlinear dynamic formulation. To achieve this goal, the dynamic-based formulation feedback linearization method is design. This method is robust and model-based nonlinear control therefore can reduce the nonlinearity term of system and reduce the effect of coupling. In this research MAMDANI fuzzy inference system is used as a main controller. It has minimum rule base to practical implementation. This technique was employed to obtain the desired control behavior with a number of information about dynamic model of system and a feedback linearization control was applied to reinforce system performance.