Journal of Advances in Computer Engineering and Technology
Volume:2 Issue: 2, Spring 2016

Pattern recognition systems are widely used in a host of different fields. Due to some reasons such as lack of knowledge about a method based on which the best classifier is detected for any arbitrary problem, and thanks to significant improvement in accuracy, researchers turn to ensemble methods in almost every task of pattern recognition. Classification as a major task in pattern recognition, have been subject to this transition. The classifier ensemble which uses a number of base classifiers is considered as meta-classifier to learn any classification problem in pattern recognition. Although some researchers think they are better than single classifiers, they will not be better if some conditions are not met. The most important condition among them is diversity of base classifiers. Generally in design of multiple classifier systems, the more diverse the results of the classifiers, the more appropriate the aggregated result. It has been shown that the necessary diversity for the ensemble can be achieved by manipulation of dataset features, manipulation of data points in dataset, different sub-samplings of dataset, and usage of different classification algorithms. We also propose a new method of creating this diversity. We use Linear Discriminant Analysis to manipulate the data points in dataset. Although the classifier ensemble produced by proposed method may not always outperform all of its base classifiers, it always possesses the diversity needed for creation of an ensemble, and consequently it always outperforms all of its base classifiers on average.

One of important aspects of software projects is estimating the cost and time required to develop projects. Nowadays, this issue has become one of the key concerns of project managers. Accurate estimation of essential effort to produce and develop software is heavily effective on success or failure of software projects and it is highly regarded as a vital factor. Failure to achieve convincing accuracy and little flexibility of current models in this field have attracted the attention of researchers in the last few years. Despite improvements to estimate effort, no agreement was obtained to select estimation model as the best one. One of effort estimation methods which is highly regarded is COCOMO. It is an extremely appropriate method to estimate effort. Although COCOMO was invented many years ago, it enjoys the effort estimation capability in software projects. Researchers have always attempted to improve the effort estimation capability in COCOMO through improving its structure. However, COCOMO results are not always satisfactory. The present study introduces a hybrid model for increasing the accuracy of COCOMO estimation. Combining bee colony algorithm with COCOMO estimation method, the proposed method obtained more efficient coefficient relative to the basic mode of COCOMO. Selecting the best coefficients maximizes the efficiency of the proposed method. The simulation results revealed the superiority of the proposed model based on MMRE and PRED(0.15).

Meta-heuristic algorithms inspired by the natural processes are part of the optimization algorithms that they have been considered in recent years, such as genetic algorithm, particle swarm optimization, ant colony optimization, Firefly algorithm. Recently, a new kind of evolutionary algorithm has been proposed that it is inspired by the human sociopolitical evolution process. This new algorithm has been called Imperialist Competitive Algorithm (ICA). The ICA is a population-based algorithm where the populations are represented by countries that are classified as colonies or imperialists. This paper is going to present a modified ICA with considerable accuracy, referred to here as ICA2. The ICA2 is tested with six well-known benchmark functions. Results show high accuracy and avoidance of local optimum traps to reach the minimum global optimal.Three important policies are in the ICA, and assimilation policy is the most important of them. This research focuses on an assimilation policy in the ICA to propose a meta-heuristic optimization algorithm for optimizing function with high accuracy and avoiding to trap in local optima rather than using original ICA by a new assimilation strategy.

Hybrid fuzzy expert systems are one of the most practical intelligent paradigm of soft computing techniques with the high potential for managing uncertainty associated to the medical diagnosis. The potential of genetic algorithm (GA) by inspiring from natural evolution as a learning and optimization technique has been vastly concentrated for improving fuzzy expert systems. In this paper, the GA capabilities have been applied for optimization of the membership function parameters in a fuzzy inference system (FIS) for diagnosing of acute lymphocytic leukemia in children. The fuzzy expert system utilizes the high interpretability of the Mamdani reasoning model to explain system results to experts in a high level and combines it with the GA optimization capability to improve its performance. The hybrid proposed Fuzzy-GA approach was implemented in Matlab software and evaluated on the real patients’ dataset. High accuracy of this system was achieved after GA tuning process with an accuracy about 98%. The results reveal the hybrid fuzzy-GA approach capability to assist computer-based diagnosis of medical experts, and consequently early diagnosis of the disease which is promising for providing suitable treatment for patients and saving more children’s lives.

Mobile ad hoc network congestion control is a significant problem. Standard mechanism for congestion control (TCP), the ability to run certain features of a wireless network, several mutations are not common. In particular, the enormous changes in the network topology and the joint nature of the wireless network. It also creates significant challenges in mobile ad hoc networks (MANET), density is one of the most important limitations that disrupts the function of the entire network, after multi-path routing can load balance in relation to the single-path routing in ad hoc networks better, so the traffic division multiple routs congestion is reduced. This study is a multi-path load balancing and congestion control based on the speed of rate control mechanism to avoid congestion in the network provides communication flows. Given such a speed control method that is consistent is that the destination node copy speed is estimated at intermediate nodes and its reflection in the In the forward direction confirmation to the sender sends a packet, therefore the rate quickly estimate The results of the simulation has been set to demonstrate that a given method better package delivery speed and expanded capacity and density to be effective checks congestion control method is better than The result traditional.

A sensor node is composed of different parts including processing units, sensor, transmitter, receiver, and security unit. There are many nodes in a sensor unit. These networks can be used for military, industrial, medicine, environmental, house, and many other applications. These nodes may be established in the lands of enemies to monitor the relations. Hence, it is important to consider conservation of communications, declaration, and key removal. The locations of nodes are not usually defined in the networks. When a secure connection is required they can be used by symmetrical or asymmetrical encodings. A node can just make secure connection, if they are in same radio range or have a common key. In dynamic wireless sensor networks compared with static networks the sensors are moveable and can be added or removed. This research makes an attempt to investigate the challenges of key management for encoding. It also tries to solve other remained problems in this field. Therefore, distribution and key management schemes supplying security and operational requirements of sensor networks are examined in fuzzy clustering and suitable protocol for key management.