Malware white-box evasion attack is a serious threat to machine learning-based malware classification models, where an attacker carefully inserts perturbations into a malware executable at a test time to evade a target model. Previous research introduced different white-box evasion attacks, namely padding and slack attacks, to craft malware adversarial samples and evaluated them based on the perturbation size and their evasion rate against a target model. However, there is a lack of insights into how the malware file structure and type affect the adversarial malware sample generation and their respective evasion rate. In this work, we provide a comprehensive empirical analysis by factoring in the malware structure and the type. Our analysis quantifies slack space availability in various sections, exploring how the slack space can influence the robustness of detection techniques. We further assess the relationship between malware type and evasion rate to understand how different types of malware respond to evasion attacks. Additionally, we explore the connection between each malware type and the corresponding slack space availability, analyzing how these structural factors influence the evasion rates during adversarial attacks. In our experiments, adversarial malware samples were generated using two different algorithms: gradient descent and iterative gradient sign method. This detailed analysis enhances our understanding of evasion dynamics of adversarial attacks across malware types and different structural characteristics of binary malware files.

With the increase in machine learning (ML) applications in different domains, incentives for deceiving these models have reached more than ever. As data is the core backbone of ML algorithms, attackers shifted their interest towards polluting the training data itself. Data credibility is at even higher risk with the rise of state-of-art research topics like open design principles, federated learning, and crowd-sourcing. Since the machine learning model depends on different stakeholders for obtaining data, there are no existing reliable automated mechanisms to verify the veracity of data from each source.Malware detection is arduous due to its malicious nature with the addition of metamorphic and polymorphic ability in the evolving samples. ML has proven to solve the zero-day malware detection problem, which is unresolved by traditional signature- based approaches. The poisoning of malware training data can allow the malware files to go undetected by the ML-based malware detectors, helping the attackers to fulfill their malicious goals. A feasibility analysis of the data poisoning threat in the malware detection domain is still lacking. Our work will focus on two major sections: training ML-based malware detectors and poisoning the training data using the label-poisoning approach. We will analyze the robustness of different machine learning models against data poisoning with varying volumes of poisoning data.

The inclusion of Internet of Things (IoT) devices is growing rapidly in all application domains. Smart Farming uses IoT devices to increase efficiency and optimize farming operations. These devices can be used in a cloud or edge computing infrastructure which can provide remote control of watering and fertilization, real time monitoring of farm conditions, and provide solutions for more sustainable practices. These improvements to efficiency and ease of use come with added risks to security and privacy. Combining vulnerable IoT devices with the critical infrastructure of the agriculture domain broadens the attack surface for adversaries. Cyberattacks in a large coordinated manner could disrupt the economy of agriculture-dependent nations. To the sensors in a system, an attack may appear as anomalous behaviour. Additionally, there are possibilities of anomalies generated due to faulty hardware, issues in network connectivity (if present), or simply abrupt changes to the environment due to weather, human error, or other unforeseen circumstances. To make these systems more secure, it is imperative to detect such data discrepancies and trigger appropriate mitigation mechanisms. In this paper, we propose an anomaly detection model for Smart Farming using an unsupervised Autoencoder machine learning model. We chose to use an Autoencoder as our method of anomaly detection because it attempts to reconstruct normal data with a low reconstruction loss and anomalous data with a high loss. The high reconstruction loss value for a data point indicates that the data is not like the rest. Our model was trained and tested on data collected from our greenhouse test-bed. Our proposed Autoencoder based anomaly detection method achieved 98.98% and took 262 seconds to train and has a detection time of .0585 seconds.

In recent years, machine learning approaches are gaining popularity with the advent of big data. The massive amount of data generated, when served as an input to machine learning approaches, provides useful insights. Adoption of these approaches in the agricultural sector has immense potential to increase crop productivity and quality. In this paper, we analyze the crop data collected from an agriculture site in Rajasthan, India, that includes both Rabi and Kharif cropping patterns. In addition, we utilize a smart farm ontology that contains concepts and properties related to the agricultural domain. We link the collected data and our smart farm ontology to populate a knowledge graph. We utilize the generated knowledge graph to provide structural information and aggregate data by using SPARQL queries. The aggregated data is further used by our machine learning models to predict the crop yield to benefit farmers and various stakeholders. We also analyze and compare our results obtained for various machine learning models used.