NASHRIYYAH -I MUHANDISI -I BARQ VA MUHANDISI -I KAMPYUTAR -I IRAN, B- MUHANDISI -I KAMPYUTAR
Year:2013 | Volume:11 | Issue:1|Papers Count:6

Learning Bayesian network structure from data has attracted a great deal of research in recent years. It is shown that finding the optimal network is an NP-hard problem when data is complete. This problem gets worse when data is incomplete i.e. contains missing values and/or hidden variables. Generally, there are two cases of learning Bayesian networks from incomplete data: in a known structure, and unknown structure. In this paper, we try to find the best parameters for a known structure by introducing the “effective parameter”, in a way that the likelihood of the network structure given the completed data being maximized. This approach can be attached to any algorithm such as SEM (structural expectation maximization) that needs the best parameters to be known to reach the optimal Bayesian network structure. We prove that the proposed method gains the optimal parameters with respect to the likelihood function. Results of applying the proposed method to some known Bayesian networks show the speed of the proposed method compared to the well-known methods.

In this paper, after a review on well-known scenario-based methods of SA evaluation, a different approach is introduced to find architectural defects. The proposed method at first, elicits the problems threatening the system's success. Then based on the analysis of the problems and probable defects which could cause the problems, tests are designed and applied to the system in order to find the real defects specially the architectural ones. Results show that the proposed method could be use to find those architectural defects which may be remained covered after applying the other methods. Therefore, it could be used as a mean to SA testing and also as a complementary mechanism along with well-known SA evaluation methods. The proposed method and its components are presented in a systematic form. An illustration of its application on the architecture of a real system is presented and the results are compared with the results of applying ATAM on the same architecture.

Texture classification is one of the main steps in image processing and computer vision applications. Feature extraction is the first step of texture classification process which plays a main role. Many approaches have proposed to classify textures since now. Among them, Local Binary Patterns and Modified Local Binary Patterns, because of simplicity and classification accuracy, have emerged as one of the most popular ones. The Local Binary Patterns have simple implementation, but with increase in the radius of neighborhood, computational complexity will be increased. Modified Local Binary Patterns assigns various labels to uniform textures and a unique label to all non-uniform ones. In this respect, the modified local binary pattern can't classify non uniform textures as well as uniform ones. In this paper a new version of Local Binary Pattern is proposed that has less computational complexity than Local Binary Patterns and more classification accuracy than Modified version. The proposed approach classifies non uniform textures as well as uniform ones. Also with change in the length of central gray level intervals, locality and globally of the features can be controlled. Classification accuracy on two standard datasets, Brodatz and Outex, indicates the efficiency of the proposed approach.

Semantic web has been one of the most important research areas of computer science in recent years. The concept of ontology as one of the most elements of semantic web is used to formally describe the domain knowledge and to enable the reasoning capability. Semantic web is a distributed system and ontologies may be developed on many different nodes, so centralized reasoning is very hard or even impossible in many cases. On the other hand, since the majority of information in semantic web is uncertain, considering the notion of uncertainty in ontological reasoning is crucial. In this paper a method for distributed reasoning in uncertain ontologies is proposed. For this purpose the distributed description logic (DDL) framework and the certainty theory are considered for distributed reasoning and modeling the uncertainty respectively. To evaluate the functionality and performance of the algorithm, we developed a case study on application of the proposed method in purifying the mappings between ontologies. The results show that our algorithm makes the mappings more precise than other similar methods.

The value of learning rate and its change mechanisms is one of the issues in designing learning systems such as learning automata. In most cases a time-based reduction function is used to adjust the learning rate aim at reaching stability in training system. So the learning rate is a parameter that determines to what extent a learning system is based on past experiences, and the impact of current events on it. This method is efficient but does not properly function in dynamic and non-stationary environments.In this paper, a new method for adaptive learning rate adjustment in learning automata is proposed. In this method, in addition to the length of time to learn, some statistical characteristics of actions probability vector of Learning Automata are used to determine the increase or decrease of learning rate. Furthermore, unlike existing methods, during the process of learning, both increase and decrease of the learning rate is done and Learning Automata responds effectively to changes in the dynamic random environment.Empirical studies show that the proposed method has more flexibility in compatibility to the non-stationary dynamic environments and get out of local maximum points and the learned values are closer to the true values.

Central Force Optimization (CFO) is a new member of heuristic algorithms which has been recently proposed and added to swarm intelligence algorithms. In this paper, an effective unsupervised image clustering technique is proposed, using CFO and called CFO-clustering. In the presented method, each probe includes the information of center of the clusters, and fitness function contains both inter-distance and intra-distance of the samples. Extensive experimental results show that the proposed CFO-clustering outperforms other similar clustering algorithms which were designed based on the evolutionary techniques.