Accepted Papers

  • ANDROID MALWARE DETECTION BASED ON MULTI-FEATURE
    Sisi Chen1,Longjiang Guo1,2,Jinbao Li1,2, Na Jiang1,and Xiaodan Guo1, 1School of Computer Science and Technology, Heilongjiang University, Harbin, China and 2Key Laboratory of Database and Parallel Computing, Heilongjiang, China
    ABSTRACT

    As smartphone usage becomes more common, the number of malware in smartphones continues to grow. Android system as a free and open source system,which has now became the most popular operating system. Aiming at the security problem of Android platform, this paper extracts various features of Android benign software and malware, and then uses CFS-GA feature selection algorithm to filter out the optimal feature subset. Finally, we use a variety of machine learning classification algorithms to evaluate malware. The experimental results show that the combination of {multi - feature and CFS- GA} reduces the feature dimension and improves the classification accuracy.

  • IMPLEMENTATION OF A SMART HOUSE APPLICATION USING WIRELESS SENSOR NETWORKS
    Ismail MOHAMMED1 and Dr. Erkan DUMAN2, P.G. Student, Department of Information Technology, Technical College of Informatics, Sulaimani Polytechnic University, Sulaimani, Iraq and 2Asst. Professor, Department of Computer Engineering, Faculty of Engineering, Firat University, 23119 Elazğ, Turkey
    ABSTRACT

    Digitisation and automation are becoming increasingly prevalent in our daily lives, both inside and outside the house. They are all attempts to simplify, use easily, monitor and be aware of any devices in a house that are connected to the system. These automated systems can be an alternative to other home manual settings. Smart homes are rarely found in my country, iraq. People’s unfamiliarity with them, the lack of use, the costs and a failure to see the importance of such systems are all possible explanations for this. Nowadays, it is generally advancing very rapidly, the internet and smart phones in particular. Using such technology and devices is inevitable and difficult to avoid. Such smart systems are able to indicate and control light, temperature, dew point, gas flow, fire ignition, opening doors and buzzing alarms. Now, as iot is emerging more widely than before, it is expected that IOT will become more prevalent and occupy wider areas of the world. IOT enables a better connection among devices in a smart home or to other smart systems via the internet and WSN. In this paper, a smart home via smart phones and pcs will be explained. Controlling and indicating electrical and electronic devices in houses and buildings will also be presented.

  • HOW TO EXTRACT WEIGHTS AND RANKING FROM FUZZY PAIRWISE COMPARISON MATRICES USING A BRAIN INSPIRED METHOD
    Mohamed Naili1,2, Abdelhak Boubetra2, Abdelkamel Tari3, Makhlouf Naili4, 1Computer science department, Faculty of Exacte Sciences, University of Bejaia, Algeria, 2Computer science department, University of BordjBouArreridj, Algeria, 3Laboratory of Medical Computing (LIMED), Faculty of Fundamental Sciences, University of Bejaia, Algeria and 4Department of Computer Science, Faculty of Exact and Natural Sciences, University of Biskra, Algeria
    ABSTRACT

    Nowadays e-commerce field is full of questions like: “How to improve my website ranking?” or “which is the best website to publish ads in?”. this kind of problems fall into what’s known as Multi-Criteria Decision Making problems in which a set of criteria has to be carefully defined and evaluated especially in case of uncertainty. To tackle this challenge, the authors of this paper present an application of a brain-inspired method, in which criteria’ weights are extracted on the basis of pair-wise comparison matrices. Afterwards, the set of criteria are used to rank a set of medical institutions’ websites.

  • DYNAMIC TASK SCHEDULING METHOD IN CLOUD COMPUTING ENVIRONMENT USING OPTIMIZED NEURAL NETWORK
    Noha Hamdy, Amal Elsayed Aboutabl and Mostafa-Sami Mostafa, Faculty of Computers and Information, Helwan University, Cairo, Egypt , Faculty of Helwan University, Computers and Information, Cairo, Egypt and Faculty of Helwan University, Computers and Information, Cairo, Egypt
    ABSTRACT

    Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources. There are two points of view to the service in the cloud from customer perspective and cloud service provider perspective. The first one focuses on minimizing response time and cost of the service and the second focuses on utilization of cloud resources and minimizing costs of maintenance. To satisfy these two points of view we have to provide an efficient way to optimize cloudlets scheduling process to available virtual machines under predefined circumstances. This paper proposes a scheduling method for cloud computing environment based on artificial neural network (ANN) optimized with firefly algorithm to pick out the most convenient scheduling algorithm