نشریه فناوری اطلاعات و ارتباطات ایران
سال هشتم شماره 27 (بهار و تابستان 1395)

Texture analysis plays an important role in image processing. Considering the extraordinary appearance texture sonar images, texture analysis are good choices for analysis of acoustic seabed images. Local binary pattern (LBP) operator is a very efficient and multi-resolution texture descriptor. It acquires appropriate information from the illumination and moods of images. Despite many developing of the LBP have proposed, but they are sensitive to noise. Also sometimes they lead to describe different structural patterns with same binary codes, which would reduce their ability to differentiate. This paper proposes an overview in provided LBP methods which includes several of the newer ones. Then it proposes a robust framework of binary pattern as completed robust LBP to overcome the inefficiency of all types of LBP which the value of the central pixel replace whit average value of intensity values of 3*3 square adjacent. Proposed approach is a quick tool with high accuracy in the classification of the seabed images. The comparison of simulation results with other known methods indicates the effectiveness of the proposed algorithm.

In general, Spam is related to satisfy or not satisfy the client and isn’t related to the content of the client’s email. According to this definition, problems arise in the field of marketing and advertising for example, it is possible that some of the advertising emails become spam for some users, and not spam for others. To deal with this problem, many researchers design an anti-spam based on personal profiles. Normally machine learning methods for spam classification with good accuracy are used. However, there isn’t a unique successful way based on Electronic Commerce approach. In this paper, at first were prepared a new profile that can lead to better simulations of user’s behavior. Then we gave this profile with advertising emails to students and collected their answers. In continue, were examined famous methods for email classification. Finally, comparing different methods by criteria of data mining standards, it can be shown that neural network method has the best accuracy for various data sets.

Nowadays, smart and fast services to patients and move on to the next generation hospital are the essential parts of the health field. The production of information technology architecture for hospital organizations is the foundation of accessible smart services and providing services with more speed and higher quality than the traditional systems. In this paper, the target is to present an indigenous IT architecture based on important criteria and metrics in Iranian hospitals. The TOGAf architecture is used and then it is adopted and localized for Shariati hospital based on Iran’s indigenous conditions. For this aim, expert’s views and 134 questionnaire tools are applied and the results are analyzed through suitable statistical tests. So IT architecture is designed by consisting up a conceptual model of four input and four infrastructure layer. The results show that 111 items are verified to apply to the hospital among 145 possible items. The customization framework is called the Hospital IT Architecture. The components provide the framework approved in 8 layers and 11 components can be used in hospitals to implement enterprise architecture. The proposed architecture in 8 layers and 11 components are designed and it could be applied as an indigenous reference architecture for implementation of enterprise architecture in Iranian Hospital organizations.

The proposed framework applies two game theoretic models for economic deployment of intrusion detection system (IDS). The first scheme models and analyzes the interaction behaviors of between an attacker and intrusion detection agent within a non-cooperative game, and then the security risk value is derived from the mixed strategy Nash equilibrium. The second scheme uses the security risk value to compute the Shapley value of intrusion detection agent under the various threat levels. Therefore, the fair agent allocation creates a minimum set of IDS deployment costs. Numerical examples show that the network administrator can quantitatively evaluate the security risk of each intrusion detection agent and easily select the most effective IDS agent deployment to meet the various threat levels.

In order to evaluate the performance and desirability of the activities of its units each organization needs an evaluation system to assess this desirability and it is more important for financial institutions, including information technology-based companies. Data envelopment analysis (DEA) is a non-parametric method to measure the effectiveness and efficiency of decision-making units (DMUs). On the other hand, data mining technique allows DMUs to explore and discover meaningful information, which had previously been hidden in large databases. . This paper presents a general framework for combining DEA and regression tree for evaluating the effectiveness and efficiency of the DMUs. Resulting hybrid model is a set of rules that can be used by policy makers to discover reasons behind efficient and inefficient DMUs. Using the proposed method for examining factors related to productivity, a sample of 18 branches of Iran insurance in Tehran was elected as a case study. After modeling based on advanced model the input oriented LVM model with weak disposability in data envelopment analysis was calculated using undesirable output, and by use of decision tree technique deals with extracting and discovering the rules for the cause of increased productivity and reduced productivity.

Social networks are one of the types of social media which in recent years has been influential in the lives of internet users so that forms of social communication is influenced by the new media and in the future will play a more important role. On the other hand the trust is one of the major causes of satisfaction and repeat use of the online service.This research employs elaboration likelihood model to identify the factors affecting the users’ initial trust in Facebook and it express that users change their attitude via a dual route including central route and peripheral route. Central cues include information quality and service quality, whereas peripheral cues include system quality, structural assurance and reputation. Self-efficacy moderates the effects of central cues and peripheral cues on initial trust. The populations of this study are Iranian users on Facebook. Data for this study through contacts with personal pages and pages of general (relating to groups, cities, activities, university, etc.) active in the social network were collected. For this purpose, link of online questionnaire to provide them on the website docs.google.com. In this questionnaire was used a five -point Likert scale from strongly disagree (1) to strongly agree (5) and finally 406 questionnaires were collected according to schedule and Morgan Jersey, which are well to conduct research in the infinite. Questionnaire validity by the opinions of experts and university professors and its reliability through Cronbach's alpha were provided and represents optimum reliability. Data were analyzed with Smart PLS statistical software. According to the results of the independent variables in this study, self-efficacy, reputation and structural assurance respectively with path coefficient 0.47, 0.23 and 0.2 have the greatest effect on the initial trust and system quality has minimal impact on initial trust with amount 0.07. The results also indicate that self-efficacy moderates the effect of system quality on initial trust. Therefore, since the self-efficacy apart from the moderating role directly also has the greatest impact on initial trust; social network designers should be a special attention to this factor so that it's learn is easy and it's use process is clear and understandable for users.