Abstract Electronic commerce is often touted as a global phenomenon that has known its growth especially in the developed countries. However, developing countries in general and Tunisia in particular have fallen far below expectations. The noticed statistics don’t overtake a simple regulation of bills. The research in hand tends to put the accent on variables that are expected to explain the... Electronic commerce is often touted as a global phenomenon that has known its growth especially in the developed countries. However, developing countries in general and Tunisia in particular have fallen far below expectations. The noticed statistics don’t overtake a simple regulation of bills. The research in hand tends to put the accent on variables that are expected to explain the e-commerce adoption. Using dimensions of national culture and attitude as theoretical bases, this research investigates the weight of culture in the explanation of individual’s behavior concerning shopping from the site. Results indicate at first glance that both Tunisian and French samples present relatively the same rate of cultural importance in their attitude towards shopping from the site (28.3% vs. 20.7%). Secondly, similar to literature, the attitude is the predictor of the intention to shop from the site and the intention succeeds to be the intermediary variable between attitude and behavior especially for the Tunisian sample. As a consequence, when the attitude is positive, it will be transformed in an intention which in turn will be translated into a behavior. Read More Read Less

Abstract Object detection has gained a great deal of attention due to the requirements in many real world applications. Adaboost based object detector have shown good results in terms of accuracy and speed. It was not until this method that object detection became widely used in real world applications. The proposed Adaboost based detectors works well, though challenges and difficulties still... Object detection has gained a great deal of attention due to the requirements in many real world applications. Adaboost based object detector have shown good results in terms of accuracy and speed. It was not until this method that object detection became widely used in real world applications. The proposed Adaboost based detectors works well, though challenges and difficulties still remain, most of which are mainly related to a large number of examples of the training sets and long training process. A large number of features are required to be selected for Adaboost which slows down training process. In this work, we make headway toward reducing the number of features. The Genetic Adaboost method is proposed to select the most relevant features and discard redundant features. The number of features for a cascade structure was reduced to 57% compared to a standard Adaboost face detector. Read More Read Less

Abstract Because of their exceptional qualities, WSNs are grabbing attention in the research community. WSNs are very less into view. Subsequently it is easily possible for an assaulter to catch a node physically, altering its code and getting private data like cryptographic keys. Wireless medium is inherently broadcast in nature which makes it open to attacks. These attacks can bother the... Because of their exceptional qualities, WSNs are grabbing attention in the research community. WSNs are very less into view. Subsequently it is easily possible for an assaulter to catch a node physically, altering its code and getting private data like cryptographic keys. Wireless medium is inherently broadcast in nature which makes it open to attacks. These attacks can bother the operation of WSN and can even kill the purpose of their deployment [3]. Keeping in mind the end goal to ensure the data in WSN, some approaches are severely required that can make data transmission more secure from the attackers. Node replication attack or clone attack is a standout amongst the most destructive and perilous risk to an unattended wireless sensor network in light of the fact that in this attack, an adversary not only compromises the sensor nodes but can also carry out a large class of internal attacks for instance DoS attack, Sybil attack, and Black hole, and wormhole attack. Numerous strategies have been proposed for the detection of node replication attack and some fairly serious limitations like the base station introduce a single point of failure, and any compromise of the base station makes the solution useless. This leads to the motivation to work in this research area. A solution to avoid the node replication attack has been proposed. The solution will be implemented in NS-2. Read More Read Less