There are an increasing number of Internet of Things (IoT) devices connected to the network these days, and due to the advancement in technology, the security threads and cyberattacks, such as botnets, are emerging and evolving rapidly with high-risk attacks. These attacks disrupt IoT transition by disrupting networks and services for IoT devices. Many recent studies have proposed ML and DL techniques for detecting and classifying botnet attacks in the IoT environment. This study proposes machine learning methods for classifying binary classes. This purpose is served by using the publicly available dataset UNSW-NB15. This dataset resolved a class imbalance problem using the SMOTE-OverSampling technique. A complete machine learning pipeline was proposed, including exploratory data analysis, which provides detailed insights into the data, followed by preprocessing. During this process, the data passes through six fundamental steps. A decision tree, an XgBoost model, and a logistic regression model are proposed, trained, tested, and evaluated on the dataset. In addition to model accuracy, F1-score, recall, and precision are also considered. Based on all experiments, it is concluded that the decision tree outperformed with 94% test accuracy.

Nowadays, in a world full of uncertainties and the threat of digital and cyber-attacks, blockchain technology is one of the major critical developments playing a vital role in the creative professional world. Along with energy, finance, governance, etc., the healthcare sector is one of the most prominent areas where blockchain technology is being used. We all are aware that data constitute our wealth and our currency; vulnerability and security become even more significant and a vital point of concern for healthcare. Recent cyberattacks have raised the questions of planning, requirement, and implementation to develop more cyber-secure models. This paper is based on a blockchain that classifies network participants into clusters and preserves a single copy of the blockchain for every cluster. The paper introduces a novel blockchain mechanism for secure healthcare sector data management, which reduces the communicational and computational overhead costs compared to the existing bitcoin network and the lightweight blockchain architecture. The paper also discusses how the proposed design can be utilized to address the recognized threats. The experimental results show that, as the number of nodes rises, the suggested architecture speeds up ledger updates by 63% and reduces network traffic by 10 times.

Unmanned Aerial Vehicles (UAVs), or drones, provided with camera sensors enable improved situational awareness of several emergency responses and disaster management applications, as they can function from remote and complex accessing regions. The UAVs can be utilized for several application areas which can hold sensitive data, which necessitates secure processing using image encryption approaches. At the same time, UAVs can be embedded in the latest technologies and deep learning (DL) models for disaster monitoring areas such as floods, collapsed buildings, or fires for faster mitigation of its impacts on the environment and human population. This study develops an Artificial Intelligence-based Secure Communication and Classification for Drone-Enabled Emergency Monitoring Systems (AISCC-DE2MS). The proposed AISCC-DE2MS technique majorly employs encryption and classification models for emergency disaster monitoring situations. The AISCC-DE2MS model follows a two-stage process: encryption and image classification. At the initial stage, the AISCC-DE2MS model employs an artificial gorilla troops optimizer (AGTO) algorithm with an ECC-Based ElGamal Encryption technique to accomplish security. For emergency situation classification, the AISCC-DE2MS model encompasses a densely connected network (DenseNet) feature extraction, penguin search optimization (PESO) based hyperparameter tuning, and long short-term memory (LSTM)-based classification. The design of the AGTO-based optimal key generation and PESO-based hyperparameter tuning demonstrate the novelty of our work. The simulation analysis of the AISCC-DE2MS model is tested using the AIDER dataset and the results demonstrate the improved performance of the AISCC-DE2MS model in terms of different measures.

This rapidly changing digital world is always sensitive to improving security and resilience to protect the inhabitants of this ecosystem in terms of data, processes, repositories, communication, and functions. The transformation of this digital ecosystem is heavily dependent on cloud computing, as it is becoming the global platform for individuals, corporates, and even governments. Therefore, the concerns related to security are now linked closely with cloud computing. In this paper, a multi-cloud security framework takes a view on the development of security mechanisms to provide a diversion to the attacker. The purpose is to gain more time to analyze the attack and mitigate the intrusion without compromises. This mechanism is designed using the honeypot technology that has been around for some time but has not been used in cloud computing and other technologies. The proposed framework provides modules related to managing the multi-cloud platform, the intrusion detection and prevention system, and honeypots. The results show significant improvement in the accuracy of detecting attacks. These results are generated in a two-phase scenario, and the first phase has been analyzed without the engagement of the honeypot module presented in the framework. The second phase has been executed with same parameters and conditions by engaging the honeypot module. It includes a comparison taxonomy of both results and an in-depth study of existing honeypots, as well as critical design elements for current honeypot research and outstanding concerns for future honeypots in IoT, multi-cloud contexts.

The digital transformation disrupts the various professional domains in different ways, though one aspect is common: the unified platform known as cloud computing. Corporate solutions, IoT systems, analytics, business intelligence, and numerous tools, solutions and systems use cloud computing as a global platform. The migrations to the cloud are increasing, causing it to face new challenges and complexities. One of the essential segments is related to data storage. Data storage on the cloud is neither simplistic nor conventional; rather, it is becoming more and more complex due to the versatility and volume of data. The inspiration of this research is based on the development of a framework that can provide a comprehensive solution for cloud computing storage in terms of replication, and instead of using formal recovery channels, erasure coding has been proposed for this framework, which in the past proved itself as a trustworthy mechanism for the job. The proposed framework provides a hybrid approach to combine the benefits of replication and erasure coding to attain the optimal solution for storage, specifically focused on reliability and recovery. Learning and training mechanisms were developed to provide dynamic structure building in the future and test the data model. RAID architecture is used to formulate different configurations for the experiments. RAID-1 to RAID-6 are divided into two groups, with RAID-1 to 4 in the first group while RAID-5 and 6 are in the second group, further categorized based on FTT, parity, failure range and capacity. Reliability and recovery are evaluated on the rest of the data on the server side, and for the data in transit at the virtual level. The overall results show the significant impact of the proposed hybrid framework on cloud storage performance. RAID-6c at the server side came out as the best configuration for optimal performance. The mirroring for replication using RAID-6 and erasure coding for recovery work in complete coherence provide good results for the current framework while highlighting the interesting and challenging paths for future research

In recent years, with the deterioration of the earth’s ecological environment, the survival of birds has been more threatened. To protect birds and the diversity of species on earth, it is urgent to build an automatic bird image recognition system. Therefore, this paper assesses the performance of traditional machine learning and deep learning models on image recognition. Also, the help-ability of transfer learning in the field of image recognition is tested to evaluate the best model for bird recognition systems. Three groups of classifiers for bird recognition were constructed, namely, classifiers based on the traditional machine learning algorithms, convolutional neural networks, and transfer learning-based convolutional neural networks. After experiments, these three classifiers showed significant differences in the classification effect on the Kaggle-180-birds dataset. The experimental results finally prove that deep learning is more effective than traditional machine learning algorithms in image recognition as the number of bird species increases. Besides, the obtained results show that when the sample data is small, transfer learning can help the deep neural network classifier to improve classification accuracy.