Wide Web as one of the greatest achievements of technology, the economy, and more substantially influenced modern societies. ¬ The present situation is not satisfactory because many of the essential tasks of daily living are not supported by automated tools. Most of today's Web content designed for human use, while ¬ machines are only capable peers and manipulates data in the dictionary. In this regard, the concept of the Semantic Web has been proposed that the data be ¬ computers is incomprehensible. Help ontology semantic technologies such as machine can understand and be smart. Powerful tool to show and express knowledge in a specific domain ontology, the format is informal and can be processed by machines. Following this finding correspondences between ontology entities using machine learning techniques. This similarity indices calculated at various levels of the ontology pairs, then sets of data alignment is to classify products .

Various methods have been proposed to solve the traveling salesman problem such as a solution based on genetic algorithm. But in general, these algorithms and especially genetic algorithm are often faced with early convergence in solving combinatorial problem, in this paper we review and introduce new methods and survey to solve the problem. Our idea based on hybrid genetic algorithm for solving the traveling salesman problem. The implementation of our algorithm is that different methods at each step of the algorithm, including initial population by the nearest neighbor, greedy algorithm and combination of them with the genetic algorithm and also different methods for mutation operator has been surveyed. Combining untwist algorithm with genetic algorithm to improve the population and is expected to strengthen the global optimal solution. When the diversity of the population is reduced, artificial chromosomes with high diversity level are introduce to increase the diversity level in the system .Implementation has been shown the improvement result of hybrid algorithm than the pure genetic algorithm .

Today information is the most important asset and strategic resources of any organization. For this reason, information technology management, was one of the organizational planning pillars programs need to ensure the value of information technology, risk management and control information associated with it, as a key element to leadership in any organization. In this article we are going through various frameworks, COBIT, Risk IT, and Val IT frameworks to examine. business and IT are Two pillars of the organizations, To choice Framework for IT management we should consider alignment of business and goals. The framework provided help to the senior management in a framework for controlling alignment business architecture and information systems to the objectives of the organization that it was the outcome frameworks COBIT, Risk IT, and Val IT, such as investment Technology in the desired position, risk management, life cycle forecast establish a common language between IT and business executives and IT managers, validation and evaluation activities to maximize profits. COBIT have some bugs, to cover this Defect we suggest that use Val IT and Risk IT parallel to COBIT, in fact those framework are Complementary with COBIT that can to form IT Governance in Organization in conclusion Alignment goals and business .

Cluster is one of the lifetime increase of wsn. The number and the size of cluster and cluster head selection is the main factor. Increase of cluster size causes the increase of distance and the early discharge. Decrease of cluster size causes the increase of sending of overhead controlling information. Inappropriate selection of the cluster increases the distance to the member of cluster and other cluster heads. In this article, we follow the increase of lifetime in angle clustering using Three cluster heads in wsn. Cluster head selection in this methods is related to remained energy, the average distance to the members and distance between current cluster head and previous one. This way, the new cluster head and the previous one does change, if the continuous working time becomes lower than optimum value. The continuous working time of cluster head is related to the optimum one hop distance and angle clustering. Simulation result reveals that above methods have caused the lifetime increasing of the wsn.

With increasing growth in the users’ demand for high throughput, more and more energy is consumed in Wireless Mesh Networks (WMNs). On the other hand, energy efficiency is an important issue in contexts of using the energy-constrained resources or having an eco-friendly network. Since network throughput is correlated to the rate at which users send their data, we can reach to energy efficiency target in WMNs by controlling the users’ rate. Motivated by this, here we present an approach to control the users’ rate. The proposed approach is formulated by using the linear optimization framework that its objective is maximization of throughput subject to the required constraints on network lifetime and devices’ function time. Also, we formulate the problem of users’ rate control as Multi-Objective Linear Programming (MOLP) to find a tradeoff between maximizing the throughput and minimizing the aggregated power consumption of relay Wireless Mesh Routers (WMRs). Our numerical results show the effect of devices’ initial energy and function time on the network throughput. Also, our experiments illustrate a tradeoff relationship between the network throughput and lifetime, and also, the linear relationship between network throughput and aggregated power consumption of relay WMRs .

In Present Article Primarily Some Explanation About URIs In Order To Encrypting Of Physical And Virtual Resource With RDF Content Is Presented. Encrypting in the present scheme is in the form of reversible and offline and it also has advantages and disadvantages which in this article will be dealt with. the mentioned scheme is similar to URL, but with due to its disadvantages, the potential of its usage will be decreased . Next, the representation of new method for security of URL will be concerned. in this new method, encrypting of physical and virtual resource which will be presented in the form of irreversible algorithm and offline, can increase the security of URL in high extend, and it makes the host information and the wanted website not allow to hackers .

Radio frequency identification (RFID) is a remote identification technique that causes a change in Mode of using a specific object to identify in industry. However, should be considered arrangements for protection against malware threats and users' information privacy and undetectablity of RFID. Malwares are malicious software in which their main purpose is destruction of computer systems. Although the important information is stored in backend database, they are threatened by SQL injection and virus attacks while using RFID electronic tags. At first the paper introduces RFID technology, and then be subjected to analysis in order to the effective diagnostic method of automatic SQL injection prevent the system away from contamination and also Framework to protect privacy of tag information and RFID system protection against malwares, Hence by using this framework, RFID system be protected from malicious software and tag's privacy protection will be insured. .

Along with the development of content and text filtering systems, identifying and highlighting Quranic verses inside texts has been considered by researchers for years. Therefore we have designed and implemented an intelligent system that can detect and highlight Quranic verses in various texts automatically. This system is an information retrieval system and it uses text mining and pattern matching algorithms. Because of Quranic spelling variations and differences, there are many challenges in this system, so that makes it beyond the usual pattern matching systems. The main algorithm of this system is a new index-based multiple pattern matching algorithm that its main idea is mapping the text and patterns to numeric arrays and process on the numbers instead of text. Experimental results show that processing on numerical values rather than text has a significant impact on increasing the performance of algorithm to be faster and more precise .

Handwriting is used by analyzers to detect personality of an individual, but this task is time consuming and expensive. Therefore computer aided techniques can be used. The science of study and analysis of human personality based on his handwriting is named graphology. In addition, any handwriting has many features. This paper in order to graphology analyses is focused on the points detection in Farsi handwriting.According to the basic features of handwriting image such as pen width, size of words and base line, we propose three methods to detect points: detecting based on the area values, detecting based on the location of points or combination of these two properties. Proposed methods tested on various Farsi handwriting images. Results show third method which is based on the area of points and the location of points is more accurate than other methods .

There are two model for learning in machine learning: symbolic learning and sub-symbolic learning. decision tree is a symbolic approach that is very comprehensible and interpretable. neural network is sub-symbolic approach that is very stable and flexible. neural network tree is a hybrid learning model with the overall structure being a decision tree and each non-terminal node containing a neural network that try to combine advantage of decision tree and neural network. Experiment result shows that, in general, neural network tree is better than traditional decision tree, because neural network is better than one feature. .