Learning With Imbalanced Data in Smart Manufacturing: A Comparative Analysis

The explosive proliferation of Internet of Things (IoT) devices is generating an incomprehensible amount of data. Machine learning plays an imperative role in aggregating this data and extracting valuable information for improving operational and decision-making processes. In particular, emerging machine intelligence platforms that host pre-trained machine learning models are opening up new opportunities for IoT industries. While those platforms facilitate customers to analyze IoT data and deliver faster and accurate insights, end users and machine learning service providers (MLSPs) have raised concerns regarding security and privacy of IoT data as well as the pre-trained machine learning models for certain applications such as healthcare, smart energy, etc. In this paper, we propose a cloud-assisted, privacypreserving machine learning classification scheme over encrypted data for IoT devices. Our scheme is based on a three-party model coupled with a two-stage decryption Paillier-based cryptosystem, which allows a cloud server to interact with MLSPs on behalf of the resource-constrained IoT devices in a privacy-preserving manner, and shift load of computation-intensive classification operations from them. The detailed security analysis and the extensive simulations with different key lengths and number of features and classes demonstrate that our scheme can effectively reduce the overhead for IoT devices in machine learning classification applications.

Today’s Machine Learning (ML) in a blend with Internet of Things (IoT)-based cloud applications plays a significant role in our everyday life. As indicated by Gartner’s recent study, there are around 25 billion devices and a gadget interfacing with IoT including wearables and automated vehicles to smart homes and smart cities applications. All such connected (smart) devices generate immense data that needs to be examined and analysed, to ensure that they continually learn from the available datasets and better themselves without any manual interference. This is where the prerequisite for machine learning comes into being. Several ML algorithms and techniques are introduced in a short time to easily evaluate big data measurements, increasing the IoT’s productivity. Similarly, special ML techniques, such as decision trees, clustering and neural and Bayesian networks, allow devices and gadgets to discern trends from various sources in different kinds of datasets and take appropriate decisions based on their analysis.It would really be difficult for smart devices to make smart decisions gradually without including and enforcing ML. The IoT helps to interconnect various hardware devices, such as houses, cars, electronic gadgets and other devices that are integrated with actuators, sensors and software, so that data can be collected and shared. As various organizations understand the progressive capability of the IoT, they have begun finding various obstructions they have to deliver to use it productively. Numerous organizations and businesses use ML to exploit the IoT’s latent capacity. This chapter evaluates the different methods of machine learning that deal with the challenges posed in the handling of IoT data. Note that this big data is generated through the communication of Internet of Things/smart devices, and this data stored at cloud. The taxonomy of machine learning algorithms is described in this chapter, explaining how different techniques are applied to data generated using IoT devices. It will also address the future problems of machine learning for IoT data analytics.KeywordsInternet of Things (IoT)-based cloud applicationsMachine learning techniquesArtificial intelligenceClassificationSmart era

Increasing the cardinality of categorical variables might decrease the overall performance of machine learning (ML) algorithms. This paper presents a novel computational preprocessing method to convert categorical to numerical variables ML algorithms. It uses a supervised binary classifier to extract additional context-related features from the categorical values. The method requires two hyperparameters: a threshold related to the distribution of categories in the variables and the PCA representativeness. This paper applies the proposed approach to the well-known cybersecurity NSLKDD dataset to select and convert three categorical features to numerical features. After choosing the threshold parameter, we use conditional probabilities to convert the three categorical variables into six new numerical variables. Next, we feed these numerical variables to the PCA algorithm and select the whole or partial numbers of the Principal Components (PCs). Finally, by applying binary classification with ten different classifiers, we measure the performance of the new encoder and compare it with the other 17 well-known category encoders. The new technique achieves the highest performance related to accuracy and Area Under the Curve (AUC) on high cardinality categorical variables. Also, we define the harmonic average metrics to find the best trade-off between train and test performances and prevent underfitting and overfitting. Ultimately, the number of newly created numerical variables is minimal. This data reduction improves computational processing time in Internet of things (IoT) devices connected to future networks.

In the Internet of Things (IoT) technology, there are several issues to overcome, including network security. New attacks have been developed to target vulnerabilities in IoT devices, and the Internet of Things on a huge scale will exacerbate present network threats. Machine learning is becoming increasingly used in applications such as traffic classification and intrusion detection. This paper offers a technique for detecting network fault severity and identifying Internet of Things (IoT) devices based on several attributes. By pushing information to the network area, the proposed framework captures different features from each network flow in order to identify the source of traffic and the type of traffic generated is used to detect different network attacks. Different machine learning algorithms are tested to find the better suited algorithm that deliver best results. The following are some of the examples: Random Forest (RF) algorithm, decision tree, SVM, etc. After completing the experimentation, it was found that the random forest classifier and decision tree are the best performing ML algorithms to classify the network severity with an accuracy of about 96.98% and GNB is found as the least performing machine learning model with an accuracy of about 54.41%.

As the number of Internet of Things (IoT) devices and applications increases, the capacity of the IoT access networks is considerably stressed. This can create significant performance bottlenecks in various layers of an end-to-end communication path, including the scheduling of the spectrum, the resource requirements for processing the IoT data at the Edge and/or Cloud, and the attainable delay for critical emergency scenarios. Thus, a proper classification or prediction of the time varying traffic characteristics of the IoT devices is required. However, this classification remains at large an open challenge. Most of the existing solutions are based on machine learning techniques, which nonetheless present high computational cost, whereas they are not considering the fine-grained flow characteristics of the traffic. To this end, this paper introduces the following four contributions. Firstly, we provide an extended feature set including, flow, packet and device level features to characterize the IoT devices in the context of a smart environment. Secondly, we propose a custom weighting based preprocessing algorithm to determine the importance of the data values. Thirdly, we present insights into traffic characteristics using feature selection and correlation mechanisms. Finally, we develop a two-stage learning algorithm and we demonstrate its ability to accurately categorize the IoT devices in two different datasets. The evaluation results show that the proposed learning framework achieves 99.9% accuracy for the first dataset and 99.8% accuracy for the second. Additionally, for the first dataset we achieve a precision and recall performance of 99.6% and 99.5%, while for the second dataset the precission and recall attained is of 99.6% and 99.7% respectively. These results show that our approach clearly outperforms other well-known machine learning methods. Hence, this work provides a useful model deployed in a realistic IoT scenario, where IoT traffic and devices’ profiles are predicted and classified, while facilitating the data processing in the upper layers of an end-to-end communication model.

Internet of Things (IoT) have opened the door to a world of unlimited possibilities for implementations in varied sectors in society, but it also has many challenges. One of those challenges is security and privacy. IoT devices are more susceptible to security threats and attacks. Due to constraints of the IoT devices such as area, power, memory, etc., there is a lack of security solutions that are compatible with IoT devices and applications, which is leading this world of securely connected things to the “internet of insecure things.” A promising solution to this problem is going beyond the standard or classical techniques to implementing the security solutions in the hardware of the IoT device. The integration of emerging technologies in IoT networks, such as machine learning, blockchain, fog/edge/cloud computing, and quantum computing have added more vulnerable points in the network. This paper introduces a comprehensive study on IoT security threats and solutions. Additionally, this survey outlines how emerging technologies such as machine learning and blockchain are integrated in IoT, challenges resulted from this integration, and potential solutions to these challenges. The paper utilizes the 4-layer IoT architecture as a reference to identify security issues with corresponding solutions.

Usually, IoT devices are designed to perform specific tasks, while robots have to adapt to unpredictable situations. Artificial intelligence and machine learning help these robots cope with the emerging unexpected conditions. The Internet of Robotic Things is an evolving concept that brings together all-encompassing sensors and devices with robotic and autonomous systems. Both IoT devices and robots rely on sensors to understand the surrounding environment, to process data quickly, and to decide how to respond. Nevertheless, while most IoT systems can handle only well-defined tasks, robots can handle expected situations as well. The Internet of Robotic Things is a better Internet of Things (IoT) solution due to its ability to bridge the gap between IT and real operations. The IoRT may face several challenges in power consumption and network bandwidth limitation due to the growth in the information dissemination over the Robotic network. The main objective of this paper is to introduce an overview of the concepts and challenges in the Internet of Robotic Things based on Fog Computing technique. Further, a framework of Fog based IoRT is introduced.

This paper investigates the integration of machine learning and Internet of Things (IoT) technologies inside a cloud environment in retail stores. Although IoT, ML, and big data have been studied individually, their combined usefulness and efficiency research is still lacking. To fill this gap, our research suggests a strategy that uses ML-machine learning algorithms and IoT devices to improve consumer experiences and streamline operations in retail settings. This paper collects data through Internet of Things (IoT) devices with sophisticated sensing features, such as fuzzy logic, to track and examine consumer interactions, movements, and behaviours in retail settings. Hence, Product movements can be tracked via RFID-tagged product tracking, providing information about consumer preferences and the popularity of particular products. The cloud securely stores and processes the data produced by these Internet of Things devices, using big data techniques to glean insights from enormous datasets. The most frequently used ML technique is association rule mining, which extracts significant patterns and correlations from the collected data. Retailers can use this to give personalised suggestions and enhance product placement and assortment by gaining better insights into customer preferences, purchasing patterns, and product affinities. Using case studies and realworld examples, we illustrate the effectiveness and advantages of combining IoT and machine learning technology in retail environments. This research advances the field of retail analytics by offering a thorough framework for utilising emerging technologies to promote corporate growth, improve operational efficiency, and provide better consumer experiences.

The proliferation of Internet of Things (IoT) devices in personal and household environments has led to a significant increase in security vulnerabilities. These devices, due to their limited computational resources, often struggle to support conventional security solutions, making them prime targets for cyberattacks. This paper explores the potential of lightweight machine learning (ML) models to enhance the security of personal IoT devices. By leveraging the power of AI, lightweight models can offer real-time threat detection and anomaly identification without compromising the device's performance or requiring extensive computational resources. The paper investigates various ML techniques that are well-suited for IoT environments, such as decision trees, k-nearest neighbours (KNN), and support vector machines (SVM), focusing on their ability to detect intrusions, unauthorized access, and other malicious activities while maintaining efficiency. Additionally, the study highlights the trade-offs between model complexity, accuracy, and resource consumption, offering practical insights for deploying ML solutions in resource-constrained IoT systems. Key challenges, including data privacy, model generalization, and the adaptability of models to diverse IoT ecosystems, are addressed. Finally, the paper discusses future directions for the integration of more advanced lightweight models, such as federated learning and edge computing, which could further enhance security capabilities while ensuring minimal impact on IoT device performance. Through this review, the paper advocates for the adoption of lightweight ML models as a feasible and scalable solution to securing personal IoT devices in an increasingly connected world.

The Internet of Things (IoT) is well known for being a major source of Big Data since it builds on the connection to the Internet of a huge number of smart devices that continuously report the status of their physical environments. Although the IoT paradigm is focused on the connection of objects, its real potential lies not in the objects themselves, but in the ability to generate valuable knowledge from the data extracted from these objects. It can be said that the Internet of Things is actually not about things , but about data . In this context, machine learning (ML) is a promising technique to process the generated data and transform them into information, knowledge, predict trends, produce valuable insights, and guide automated decision‐making processes. However, the use of ML techniques in IoT brings up several challenges, especially regarding the computational requirements demanded by them. Data produced by IoT devices can be typically processed in three different layers: in the data sources (Things layer), in computational clouds, and in the intermediate layer known as fog or edge. Depending on the application quality of service (QoS) parameters (such as response time) and the complexity of the required processing, each layer will be more suitable for performing a given ML technique. The overhead imposed by current ML techniques on devices of the fog and things layers, where resources are scarce, hinders their widespread adoption in this context. Recent developments in running ML algorithms on fog devices show that, due to their limited hardware and power supply, the decision‐making process in the fog is not an easy task. On the other hand, a decision‐making process built upon information extracted from several devices on the network is more reliable than by taking some limited view of the context. This chapter presents the challenges to perform big data/big stream analytics in fog computing using the recent artificial intelligence developments.

In the recent days, wide-ranging cellular devices and purchaser gadgets in the Internet of Things (IoT) have created immense multimedia information in different types of media (for example, content, pictures, video, and sound). Due to this, there is a great increase in the research challenges for creating strategies and tools in addressing Multimedia Big Data (MMBD) for future IoT. As the worldwide framework for the ongoing data society, IoT empowers progressed benefits by interconnecting (virtual as well as physical) things dependent on existing and advancing interoperable data and correspondence advancements. An immense measure of connected objects will be installed universally in a couple of years. In the meantime, the utilization of MMBD has been developing colossally since recent years, while organizations are rapidly getting on what they remain to pick up. Actually, these two advances are affecting and molding one another. In spite of the fact that they emerge from various application situations, MMBD can be together utilized with machine learning, AI, factual, and other progressed procedures, models, and techniques to investigate or locate the profound incentive behind the immense information originated from IoT. Actually, the registering knowledge, including transformative calculation, neural systems, and the fuzzy hypothesis, is relied upon to assume a vital job for these issues. It is as yet one of the most scorching and most challenging fields to create novel processing knowledge for the reasonable situations concerned with the MMBD for future IoT. In this paper, we focus on one of the most important research domains in MMBD IoT, Energy Conservation. IoT devices communicate through the wireless communication medium and are expected to transmit information whenever needed. The battery life of IoT devices is an important concern for researchers and device manufacturers. Many exhaustive efforts have been put by researchers in this area. Since most IoT devices are usually deployed in remote and hostile environments out of reach for human users, it may not be possible to charge and recharge batteries frequently. Moreover, in MMBD IoT applications, a large volume of multimedia traffic needs to be processed, which consumes precious network resources such as bandwidth and energy. Thus, devising protocols for conserving energy of IoT devices in such environments has become a very interesting topic of research. There are various ways to achieve energy conservation in the MMBD IoT environment. Some of the popular research inclinations are designing energy-efficient communication protocols, developing of mechanisms that enable IoT devices to self-generate, recycle, store and harvest energy, and modifying underlying protocol stack of communication technologies to support energy efficiency. Our paper mainly focuses on the investigation of existing technologies and mechanisms in the above domains. We first present the need for energy conservation briefly and then discuss the key points of existing solutions for saving energy in IoT communications. At the end of the paper, we summarize our findings to describe the advantages and limitations of existing mechanisms and provide insights into possible research directions.

The Internet of Things (IoT) has emerged as one of the fastest-growing areas in technology, but this growth also increases the risk of cyber-attacks on IoT devices, particularly through malware infections. In this paper, we present a novel, lightweight approach to detect IoT malware by focusing on the ELF header information, with data sizes varying from 64 to 1024 bytes. This approach aims to improve the overall precision of the results. The motivation behind this approach is to enhance security while minimizing resource usage, crucial for IoT devices with limited computational power. The ELF header is used exclusively because it provides essential metadata, enabling efficient feature extraction without the need for resource-intensive analysis of entire files. We utilize entropy and string methods, including WEM, BEM, 2-gram, and 2-gram matrix, to derive features from the ELF header, while increasing the data size. These features are then classified using machine learning algorithms such as random forest, extra trees, adaboost, and xgboost. Our method, tested on real-world datasets of malware samples targeting ARM architectures, achieves a detection accuracy of 99.98% and F1-score of 99,99% for a processing data length of 512 bytes. This fast and effective approach can be seamlessly deployed in IoT environments, enhancing security and resilience against malware infections, and contributing to the overall protection of IoT devices and networks.

With the rapid development of Internet of Things (IoT) and the use of smart devices and social networks in our daily lives, applications-based on IoT are growing exponentially in many fields such industries, business and daily life activities. The IoT technology brings a lot of promise for humanity by improving life quality and comfort and by strengthening human bonds, among others. In the next few years, billions of connected devices will be spread across smart homes, vehicles, cities, and industries. Such connected devices, with restricted resources, will interchange with users and the surrounding environment. In this context, Machine Learning (ML), Which is able to provide embedded intelligence in the IoT devices and networks, can be leveraged to decode the meaning and behavior behind the device’s data, implement accurate predictions, and make decisions for several tasks. In this chapter, we present an overview of research works about ML-base IoT systems in different areas of applications. First, we present a deep overview of IoT’s technology. Then, we highlight the most fundamental concepts of ML categories and algorithms. After that, we shed light on the ML-based IoT critical challenges and provide some potential future research directions. Eventually, we present an IoT-based ML technique scenario for smart irrigation in Agriculture 4.0.

The Internet of Things (IoT) devices is able to communicate with another IoT devices in heterogeneous environments. They are also able to create a network by their own using wireless capabilities without a centralized approach. This network is called wireless mobile ad hoc network (MANET). The MANET facilitate IoT devices to communicate with each other in an infrastructure less network of the Internet of Things environment. The IoT devices can dynamically join, create connection and exchange information among them. The role of cloud in this framework is to store and process the data by IoT devices. The communication Security has been raised as one of the most critical issues of cloud computing where resolving such an issue would result in a constant growth in the use and popularity of cloud. Our purpose of this study is to create a framework for providing the communication in an integrated framework of Cloud and MANET among IoT devices. Our main contribution links a new study that integrates cloud computing to the MANET and provides communication to the internet of things. This study can be used in the framework of the Internet of Things in future.

As deployment of Internet of Things (IoT) devices gain momentum, there is an increased interest in implementing machine learning (ML) algorithms on IoT devices. Most of the existing ML solutions, however, rely on a central server to execute data-intensive ML models, because most devices in IoT systems do not have sufficient storage and computing resources. This paper presents a distributed Artificial Neural Networks (ANN) architecture, called PCANN, which allows execution of a complex image recognition task on a collection of resource-constrained IoT devices. Our solution separates a single ML model into multiple small modules that are executed by the distributed IoT devices. The solution effectively reduces storage and computing requirements for individual devices to store and process ML model. We design multiple PCANN models and utilize the models for human posture recognition as the case study. The experimental results show that the distributed PCANN architecture achieves comparable accuracy as the classical ANN model, while the average size of each PCANN module is largely reduced.