A Stacking Ensemble Framework for Android Malware Prediction
Abstract
Every Android application needs the collection of permissions during installation time, and these can be used in permission-based malware detection. Different ensemble strategies for categorising Android malware have recently received much more attention than traditional methodologies. In this paper, classification performance of one of the primary ensemble approach (Stacking) in R libraries in context of for Android malware is proposed. The presented technique reserves both the desirable qualities of an ensemble technique, diversity, and accuracy. The proposed technique produced significantly better results in terms of categorisation accuracy.KeywordsStackingEnsembleClassificationVotingAndroid malwares
The Lack of infrastructure makes secured data distribution, challenging task in Wireless Sensor Networks (WSNs). In traditional routing methods, either security or routing optimization is addressed separately; however, both are not addressed at similar instances. Hence, if there exists a bottleneck while handling security or routing, where either one is affected by the other. In this paper, Mutual Trust Management (MTM) framework is designed between the sensor nodes is proposed in WSN to identify the malicious nodes. The trust model is connected with an Intrusion Detection System (IDS) to effectively analyse the malicious nodes and routing of packets between the nodes is designed with the Structure of a Multilayer Perceptron (MLP) Network to route the packets through the secured path. The simulations are conducted using the NS-2 setup for validating the trustworthiness and packet delivery through the secured route. The proposed method is compared against the existing methods to test the efficacy of MTM-MLP model and the results show that the MTM-MLP achieves higher detection against ransomware than the other methods.
Wireless Ad hoc Network is established by a collection of mobile nodes without any fixed infrastructure, where each node plays a role of the router. There are not any centralize control to handle the routing process of network, due to the dynamic tropology and infrastructure less network the network is vulnerable to various kinds of attacks. Therefore, numerous proactive, reactive and hybrid routing protocols have been recommended, among which one of the well-known a protocol is AODV due to its high-performance gain. This research work contributes towards mitigating network layer attacks on routing protocols in Wireless Ad hoc Networks. Problem and it's security issues because its consequences and existing mechanisms for detection and prevention with the context of AODV protocol is a challenge in Wireless Ad hoc Network, particularly in MANET and Sensor network. We present an AODV based secure routing algorithm for detection and prevention of different network layer attacks such as blackhole and rushing attacks. We use different types of security parameters like node sequence numbers, hop count, trust value, path value, acknowledge time, the threshold value and ALERT packet message to design a secure algorithm for AODV routing protocol. It shows enactment evaluation of AODV with the enhanced secure routing algorithm and existing routing algorithm through simulations which will confirm the effectiveness and accuracy of the algorithm by considering performance metrics like throughput, packet delivery ratio and end to end delay. Using network simulator NS-2.35 the experimental results have been shown an improvement in throughput, packet delivery ratio (PDR), and end to end delay using IDSAODV and results are compared with normal AODV routing protocol for blackhole and rushing attacks. The comparative results have been also shown with proposed IDSAODV and existing method.
AI plays an important role in COVID-19 identification. Computer vision and deep learning techniques can assist in determining COVID-19 infection with Chest X-ray Images. However, for the protection and respect of the privacy of patients, the hospital's specific medical-related data did not allow leakage and sharing without permission. Collecting such training data was a major challenge. To a certain extent, this has caused a lack of sufficient data samples when performing deep learning approaches to detect COVID-19. Federated Learning is an available way to address this issue. It can effectively address the issue of data silos and get a shared model without obtaining local data. In the work, we propose the use of federated learning for COVID-19 data training and deploy experiments to verify the effectiveness. And we also compare performances of four popular models (MobileNet, ResNet18, MoblieNet, and COVID-Net) with the federated learning framework and without the framework. This work aims to inspire more researches on federated learning about COVID-19.
Social Engineering is one of the easiest ways of attacking because it depends on how the
attacker can manipulate the victim. Social engineering involves getting authorized access to
secure information and locations. This form of attack is widely used, yet most organizations and
individuals still fall victim to this form of attack. Social engineering is the most challenging to
deal with because it involves humans. Organizations and prominent individuals lose a
considerable amount of money due to these attacks. Examples include sending a benign gift to
deceive the victim (trojan horse attack), sending email links that contain bugs to victims
(phishing emails), or impersonating while making a call to the victim (vishing). One way to
prevent social engineering attacks is to educate all organization staff and individuals on attackers'
different techniques to launch an attack. This paper discusses social engineering attacks, the
statistics, social engineering procedures used by attackers, and ways of mitigating these attacks.
Key terms: Footprinting, Whaling, Phishing, Quid pro quo, Tailgating (piggybacking), Dumpster
diving, Reverse social Engineering, Trojan Horses (Baiting), Ransomware, Mitigations,
Honeypot, Attacks
Binary sensing range model does not imitate the true sensing characteristics of sensor nodes (SNs) as it ignores the environmental factors affecting the sensing abilities of SNs, thus, over estimates the coverage performance. To consider the influence of environmental factors in the sensing characteristics, Elfes sensing model was proposed which is a probabilistic sensing range model. Further, sensor failure is an important issue and may cause disrupted and declined coverage and connectivity. This paper investigates the coverage performance of WSN spread in a circular region (CR) employing both Elfes and binary sensing range models. The coverage metric obtained using the derived expression employing binary sensing range model provides higher coverage as compared to coverage achieved using Elfes model. Besides, we analysed the impact of different network parameters on network coverage and found that the sensing range and the SNs count have positive impact whereas sensor failure probability and required value of k have negative impact on the networks k-coverage metric.
