Automated test generation for Andriod apps with reinforcement learning algorithms often produce test suites with redundant coverage. We looked at minimizing test suites that have already been generated based on state–action–reward–state–action (SARSA) algorithms. In this dissertation, we hypothesize that there is room for improvement by introducing novel hybrid approaches that combine SARSA-generated test suites with greedy reduction algorithms following the principle of Head-up Guidance System (HGS™) approach. In addition, we apply an empirical study on Android test suites that reveals the value of these new hybrid methods. Our novel approaches focus on post-processing test suites by applying greedy reduction algorithms. To reduce Android test suites, we utilize different coverage criteria including event-based criterion (EBC), element-based criterion (ELBC), and combinatorial-based sequences criteria (CBSC) that follow the principle of combinatorial testing to generate sequences of events and elements. The proposed criteria effectively decreased the test suites generated by SARSA and revealed a high performance in maintaining code coverage. These findings suggest that test suite reduction using these criteria is particularly well suited for SARSA-generated test suites of Android apps.

Additive manufacturing (AM) is revolutionizing the production of three-dimensional objects by converting digital design into physical forms, offering benefits such as intricate shapes, lighter products, and reduced energy consumption compared to traditional methods. However, AM faces challenges like high equipment and material costs, long printing times, and limited material variety, which hinder widespread adoption and complicated process optimization. Investment in expensive 3D printers and materials, along with printing times from hours to days, are significant obstacles to mass production. To address these challenges, machine learning offers a solution by using algorithms to create optimal models and predict material properties, thereby expediting the optimization process. In AM, complex physical reactions and cooling rates can lead to deformations and defects that impact part quality and strength. This complexity is magnified in multi-layer, multi-track printing, requiring careful monitoring of melt pool morphology and defects. Fine-grained microstructure analysis is crucial for tailoring materials to specific performance requirements. Machine learning and deep learning, through data-driven modeling, provide a rapid path and potential for optimization. This dissertation explores accelerating AM optimization and underlines the pivotal role of machine learning in overcoming the associated challenges.

Creating a robust supply chain is one of the factors for more sturdy agriculture. Most of the agricultural produce is getting wasted while storing and transporting the goods. AgroString in Section 3 system collects real-time temperature and humidity data from the IoAT edge device and performs secure data storage and transmission through a distributed ledger. Research and studies are being conducted to forecast the availability of clear groundwater with the help of traditional techniques to meet worldwide food requirements. Collecting quality data from various groundwater sites for storing and sharing for further analysis has become a more significant challenge. Our current work, Section 4, G-DaM, increases the value and reliability of groundwater data by implementing Distributed ledger with a public Blockchain, Ethereum, on the edge layer. Agriculture uses 65% of the world's freshwater for farming, half of which goes wasted; the same is the scenario for energy. We design an insurance system called IncentiveChain, which uses a distributed ledger on edge to incentivize farmers whenever they use resources at a needed level to give similar or more agricultural yield in Section 5. In the current research, we address some of the problems in data management and implement state-of-the-art distributed ledger …

The advancement in the field of artificial intelligence has been rapid in recent years and has revolutionized various industries. For example, convolutional neural networks (CNNs) perform image classification at a level equivalent to that of humans on many image datasets. These state-of-the-art networks reached unprecedented success using complex architectures with billions of parameters, numerous kernel configurations, weight initialization and regularization methods. This transitioned the models into black-box entities with little to no information on the decision-making process. This lack of transparency in decision making and started raising concerns amongst some sectors of user community such as the sectors, amongst others healthcare, finance and justice. This challenge motivated our research where we successfully produced human interpretable influential features from CNN for image classification and captured the interactions between these features by producing a concise decision tree making accurate classification decisions. The proposed methodology made use of pre-trained VGG16 with finetuning to extract feature maps produced by learnt filters. A decision tree was then induced on these extracted features that captured important interactions between the features. On the CelebA image dataset, we successfully produced human interpretable rules capturing the main facial landmarks responsible for segmenting males from females with the use of …

This research comprises three interconnected studies, all anchored in the usability evaluation of mobile education applications, with guidance from the well-established Jakob Nielsen factors to heuristic evaluation. The first study delves into the analysis of mobile application reviews using a deep learning model and machine learning to unearth usability issues. In the second study, we examine the usability of two prominent educational applications, Canvas and Blackboard, integrated within Prince Sattam bin Abdulaziz University (PSAU) and at the University of North Texas (UNT) from a student-oriented perspective. Through the synthesis of findings and insights from antecedent studies, we seek to augment the current body of knowledge and offer realistic recommendations for the enhancement of mobile education application usability. Our findings have the potential to improve the efficacy of platforms, offering developers a roadmap to refine application features and optimize the learning experience for both educators and learners.

This research delves into the application of semantic segmentation in precision agriculture, specifically targeting the automated identification and classification of various irrigation system types within agricultural landscapes using high-resolution aerial imagery. With irrigated agriculture occupying a substantial portion of US land and constituting a major freshwater user, the study's background highlights the critical need for precise water-use estimates in the face of evolving environmental challenges, the study utilizes advanced computer vision for optimal system identification. The outcomes contribute to effective water management, sustainable resource utilization, and informed decision-making for farmers and policymakers, with broader implications for environmental monitoring and land-use planning. In this geospatial evaluation research, we tackle the challenge of intraclass variability and a limited dataset. The research problem centers around optimizing the accuracy in geospatial analyses, particularly when confronted with intricate intraclass variations and constraints posed by a limited dataset. Introducing a novel approach termed "dynamic contrastive learning," this research refines the existing contrastive learning framework. Tailored modifications aim to improve the model's accuracy in classifying and segmenting geographic features accurately. Various deep learning models, including EfficientNetV2L, EfficientNetB7, ConvNeXtXLarge, ResNet-50, and ResNet-101, serve as backbones to assess their performance in the geospatial context. The data used for evaluation …

Herein I document my exploration into the intersection of convolutional neural networks and raw non-lexical audio signals by detailing the development and results of four projects, each representing a unique problem in this domain: mutation detection, upscaling, classification, and generation. Convolutional neural networks, within the class of computational models which approximate a functional relationship between spaces of data expressed through a bio-inspired structure of modular interconnected neural nodes, are a subcategory suited to data with features that are spatially correlated while variable in absolute position. Dilated convolutional neural networks are of particular interest for operating on audio signals, as the exponential dilation stack both greatly expands the receptive field and extracts features at a progression which reflects the logarithmic properties of human hearing. More generally, I seek to study at a granular level the application of convolutional neural networks to any discrete temporal signals with dense periodic features, though the primary focus is on music and components of audio composition for music and video game production.

Globalization has transformed the pharmaceutical industry into a vast, interconnected network. However, this complexity has led to inefficiencies in the supply chain, with existing ERP systems struggling to keep up due to their centralized nature, resulting in a lack of transparency and increased errors. This research proposes efficient distributed ledger-based architectures to address these challenges. In Chapter 3, a new transparent supply chain architecture is introduced to eliminate blind spots and enable stakeholders to verify the authenticity of pharmaceutical products. This system creates a secure, single source of truth for the entire lifecycle of medicines, thereby eliminating counterfeits. Chapter 4 focuses on securing the cold pharmaceutical supply chain using IoT and distributed ledgers. The proposed architecture monitors and controls environmental parameters, ensuring safe drug transport. Scalability is addressed with a novel proof of authentication (PoAh) blockchain called EasyChain. In Chapter 5, the serialization of pharmaceutical products is enhanced through digital twinning, providing an efficient and cost-effective solution while complying with regulations. This research aims to create scalable and efficient pharmaceutical supply chains, reducing counterfeits and improving overall security.

"Crowd counting" is a term used to describe the process of calculating the number of people in a given context; however, crowd counting has multiple challenges especially when images representing a given crowd span multiple cameras or images. In this thesis, we propose a crowd counting camera array and correction (CCCAC) method using a camera array of scaled, adjusted, geometrically corrected, combined, processed, and then corrected images to determine the number of people within the newly created combined crowd field. The purpose of CCCAC is to transform and combine valid regions from multiple images from different sources and order as a uniform proportioned set of images for a collage or discrete summation through a new precision counting architecture. Determining counts in this manner within normalized view (collage), results in superior counting accuracy than processing individual images and summing totals with prior models. Finally, the output from the counting model is adjusted with learned results over time to perfect the counting ability of the entire counting system itself. Results show that CCCAC crowd counting corrected and uncorrected methods perform superior to raw image processing methods.

This thesis explores the use of spatiotemporal data mining in the air quality domain to understand causes of PM2.5 air pollution. PM2.5 refers to fine particulate matter less than 2.5 microns in diameter and is a major threat to human and environmental health. A review of air quality modeling methods is provided, emphasizing data-driven modeling techniques. While data mining methods have been applied to air quality data, including temporal sequence mining algorithms, spatiotemporal sequence mining methods have not been broadly applied to study air pollution. However, air pollution is highly spatial in nature, so such methods can offer new insights into air quality. This thesis applies one such method, the Spatiotemporal Sequence Miner (STS Miner) algorithm, to air quality data from a low-cost sensor network to explore causes and trends related to PM2.5. To facilitate the use of this method, an open-source library called OpenSTSMiner is developed to implement this algorithm. Various domain results are found; for instance, low temperature and low relative humidity are strongly associated with worsening levels of air quality. Lastly, to highlight the utility of the STS Miner algorithm, a comparison is presented between STS Miner and spatial Markov chains, another spatiotemporal modeling method used in …

Many, if not all people have online social accounts (OSAs) on an online community (OC) such as Facebook (Meta), Twitter (X), Instagram (Meta), Mastodon, Nostr. OCs enable quick and easy interaction with friends, family, and even online communities to share information about. There is also a dark side to Ocs, where users with malicious intent join OC platforms with the purpose of criminal activities such as spreading fake news/information, cyberbullying, propaganda, phishing, stealing, and unjust enrichment. These criminal activities are especially concerning when harming minors. Detection and mitigation are needed to protect and help OCs and stop these criminals from harming others. Many solutions exist; however, they are typically focused on a single category of malicious intent detection rather than an all-encompassing solution. To answer this challenge, we propose the first steps of a framework for analyzing and identifying malicious intent in OCs that we refer to as malicious mntent detection framework (MIDF). MIDF is an extensible proof-of-concept that uses machine learning techniques to enable detection and mitigation. The framework will first be used to detect malicious users using solely relationships and then can be leveraged to create a suite of malicious intent vector detection models, including phishing, propaganda, scams, …

Object detection in autonomous driving systems is a critical functionality demanding precise implementation. However, existing solutions often rely on single-sensor systems, leading to insufficient data representation and diminished accuracy and speed in object detection. Our research addresses these challenges by integrating fusion-based object detection frameworks and augmentation techniques, incorporating both camera and LiDAR sensor data. Firstly, we introduce Sniffer Faster R-CNN (SFR-CNN), a novel fusion framework that enhances regional proposal generation by refining proposals from both LiDAR and image-based sources, thereby accelerating detection speed. Secondly, we propose Sniffer Faster R-CNN++, a late fusion network that integrates pre-trained single-modality detectors, improving detection accuracy while reducing computational complexity. Our approach employs enhanced proposal refinement algorithms to enhance the detection of distant objects, resulting in significant improvements in accuracy on challenging datasets like KITTI and nuScenes. Finally, to address the sparsity inherent in LiDAR data, we introduce a novel method that generates virtual LiDAR points from camera images, augmented with semantic labels to detect sparsely distributed and occluded objects effectively and integration of distance-aware data augmentation (DADA) further enhances the model's ability to recognize distant objects, leading to significant improvements in detection accuracy overall.

Lite-Agro studies applications of TinyML in pear (Pyrus communis) tree disease identification and explores hardware implementations with an ESP32 microcontroller. The study works with the DiaMOS Pear Dataset to learn through image analysis whether the leaf is healthy or not, and classifies it according to curl, healthy, spot or slug categories. The system is designed as a low cost and light-duty computing detection edge solution that compares models such as InceptionV3, XceptionV3, EfficientNetB0, and MobileNetV2. This work also researches integration with federated learning frameworks and provides an introduction to federated averaging algorithms.

This dissertation presents novel approaches to optimizing convolutional neural network (CNN) architectures for connected autonomous vehicle (CAV) workload on edge, tailored to surmount the challenges inherent in cooperative perception under the stringent resource constraints of edge devices (an endpoint on the network, the interface between the data center and the real world). Employing a modular methodology, this research utilizes the insights from granular examination of CAV perception workloads on edge platforms, identifying and analyzing critical bottlenecks. Through memory contention-aware neural architecture search (NAS), coupled with multi-objective optimization (MOO) and the Non-dominated Sorting Genetic Algorithm II (NSGA-II), this work dynamically optimizes CNN architectures, focusing on reducing memory cost, layer configuration and parameter optimization to reach set hardware constraints whilst maintaining a target precision performance. The results of this exploration are significant, achieving a 63% reduction in memory usage while maintaining a precision rate above 80% for CAV relevant object classes. This dissertation makes novel contributions to the field of edge computing in CAVs, offering a scalable and automated pipeline framework for dynamically obtaining an optimized model for given constraints, thus enabling CAV workloads on edge. In future research, this dissertation also opens multiple different venues for areas of integration. The modular …

Software quality attributes, such as reliability, security, and usability, are often well-defined and understood at the requirement level. They lay the ground foundation necessary to achieve high-quality, robust, user-friendly, and trustworthy software systems. However, when addressing these attributes at the code level, two significant challenges emerge. First, they tend to scatter across the codebase due to improper encapsulation of object-oriented classes, hampering the visibility of quality-related components across the codebase. Second, they become tangled within a single module due to intricate interdependencies with functional aspects of the code. Addressing quality concerns in the presence of scattered and tangled code can lead to unforeseen issues. For example, software developers may inadvertently introduce new and latent bugs or incorrectly implement code components deviating from the original system-wide requirements. To tackle these pressing issues, this dissertation proposes a series of state-of-the-art solutions integrating ML-based techniques and NLP-based techniques, including static program analysis techniques, to automatically and effectively detect and repair quality concerns present at the code level, even when scattered across the codebase. Additionally, we introduce program structural analysis and change impact analysis, complemented by other unsupervised ML-based techniques, to disentangle quality-related changes from functional ones, to gain a holistic understanding of a …

This study introduces an effective method to address the generalization challenge in pretrained language models (PLMs), which affects their performance on diverse linguistic data beyond their training scope. Improving PLMs' adaptability to out-of-distribution (OOD) data is essential for their reliability and practical utility in real-world applications. Furthermore, we address the ethical imperative of fairness in PLMs, particularly as they become integral to decision-making in sensitive societal sectors. We introduce gender-tuning, to identify and disrupt gender-related biases in training data. This method perturbs gendered terms, replacing them to break associations with other words. Gender-tuning stands as a practical, ethical intervention against gender bias in PLMs. Finally, we present FairAgent, a novel framework designed to imbue small language models (SLMs) with fairness, drawing on the knowledge of large language models (LLMs) without incurring the latter's computational costs. FairAgent operates by enabling SLMs to consult with LLMs, harnessing their vast knowledge to guide the generation of less biased content. This dynamic system not only detects bias in SLM responses but also generates prompts to correct it, accumulating effective prompts for future use. Over time, SLMs become increasingly adept at producing fair responses, enhancing both computational efficiency and fairness in AI-driven interactions.

Side channels give attackers the opportunity to reveal private information without accessing it directly. In this study, several novel approaches are presented to mitigate cache side channel attacks including Spectre attack and its variants, resulting in several contributions. CHASM shows the information leakage in several new cache mapping schemes, where different cache address mappings may provide higher or lower protection against cache side channel attacks. GuardCache creates a noisy cache side-channel, making it more difficult for the attacker to determine if an access is a hit or miss (which is the basis for most side channel attacks). SecurityCloak is a framework that encompasses GuardCache with SafeLoadOnMiss whereby cache load misses during speculative execution are delayed until the speculation is resolved, thus preventing attacks that rely on accessing data in during (mis) speculated executions. To search for a compromise between security and performance, it is recommended not always to use protections such as SecurityCloak protections, but also to activate the protection only while executing critical sections of code or on-demand when an attack is detected (or suspected). Our experimental results show a high degree of obfuscation (and prevention of side channels) with a minimal impact on the performance.

This research proposes FruitPAL and FruitPAL 2.0. They are full automatic devices that can detect fruit consumption to reduce the risk of disease. Allergies to fruits can seriously impair the immune system. A novel device (FruitPAL) detecting fruit that can cause allergies is proposed in this thesis. The device can detect fifteen types of fruit and alert the caregiver when an allergic reaction may have happened. The YOLOv8 model is employed to enhance accuracy and response time in detecting dangers. The notification will be transmitted to the mobile device through the cloud, as it is a commonly utilized medium. The proposed device can detect the fruit with an overall precision of 86%. FruitPAL 2.0 is envisioned as a device that encourages people to consume fruit. Fruits contain a variety of essential nutrients that contribute to the general health of the human body. FruitPAL 2.0 is capable of analyzing the consumed fruit and then determining its nutritional value. FruitPAL 2.0 has been trained on YOLOv5 V6.0. FruitPAL 2.0 has an overall precision of 90% in detecting the fruit. The purpose of this study is to encourage fruit consumption unless it causes illness. Even though fruit plays an important role in people's …

In recent years, there have been incidents which decreased people's trust in some organizations and authorities responsible for ratings and accreditation. For a few prominent examples, there was a security breach at Equifax (2017), misconduct was found in the Standard & Poor's Ratings Services (2015), and the Accrediting Council for Independent Colleges and Schools (2022) validated some of the low-performing schools as delivering higher standards than they actually were. A natural solution to these types of issues is to decentralize the relevant trust management processes using blockchain technologies. The research problems which are tackled in this thesis consider the issue of trust in reputation for assessment and review credibility at different angles, in the context of blockchain applications. We first explored the following questions. How can we trust courses in one college to provide students with the type and level of knowledge which is needed in a specific workplace? Micro-accreditation on a blockchain was our solution, including using a peer-review system to determine the rigor of a course (through a consensus). Rigor is the level of difficulty in regard to a student's expected level of knowledge. Currently, we make assumptions about the quality and rigor of what is learned, but …

Radio map estimation (RME) is the task of predicting radio power at all locations in a two-dimensional area and at all frequencies in a given band. This thesis explores four deep learning approaches to RME: dual path autoencoders, skip connection autoencoders, diffusion, and joint learning with transmitter localization.

In this dissertation, I investigate the existing smart contract problems that limit cognitive abilities. I use Taylor's serious expansion, polynomial equation, and fraction-based computations to overcome the limitations of calculations in smart contracts. To prove the hypothesis, I use these mathematical models to compute complex operations of naive Bayes, linear regression, decision trees, and neural network algorithms on Ethereum public test networks. The smart contracts achieve 95\% prediction accuracy compared to traditional programming language models, proving the soundness of the numerical derivations. Many non-real-time applications can use our solution for trusted and secure prediction services.

In this dissertation, we address the problem of vagueness in traditional legal contracts by presenting novel methodologies that aid in the paradigm shift from traditional legal contracts to smart contracts. We discuss key enabling technologies that assist in converting the traditional natural language legal contract, which is full of vague words, phrases, and sentences to the blockchain-based precise smart contract, including metrics evaluation during our conversion experiment. To address the challenge of this contract-transformation process, we propose four novel proof-of-concept approaches that take vagueness and different possible interpretations into significant consideration, where we experiment with popular vendors' existing vague legal contracts. We show through experiments that our proposed methodologies are able to study the degree of vagueness in every interpretation and demonstrate which vendor's translated-smart contract can be more accurate, optimized, and have a lesser degree of vagueness. We also incorporated the method of fuzzy logic inside the blockchain-based smart contract, to successfully model the semantics of linguistic expressions. Our experiments and results show that the smart contract with the higher degrees of truth can be very complex technically but more accurate at the same time. By using fuzzy logic inside a smart contract, it becomes easier to solve the …

This dissertation introduces a hybrid strategy for automated testing of Android applications that combines reinforcement learning and test suite prioritization. These approaches aim to improve the effectiveness of the testing process by employing reinforcement learning algorithms, namely Q-learning and SARSA (State-Action-Reward-State-Action), for automated test case generation. The studies provide compelling evidence that reinforcement learning techniques hold great potential in generating test cases that consistently achieve high code coverage; however, the generated test cases may not always be in the optimal order. In this study, novel test case prioritization methods are developed, leveraging pairwise event interactions coverage, application state coverage, and application activity coverage, so as to optimize the rates of code coverage specifically for SARSA-generated test cases. Additionally, test suite prioritization techniques are introduced based on UI element coverage, test case cost, and test case complexity to further enhance the ordering of SARSA-generated test cases. Empirical investigations demonstrate that applying the proposed test suite prioritization techniques to the test suites generated by the reinforcement learning algorithm SARSA improved the rates of code coverage over original orderings and random orderings of test cases.

The original FlexiChain and its descendants are a revolutionary distributed ledger technology (DLT) for cyber-physical systems (CPS) and their embedded systems (ES). FlexiChain, a DLT implementation, uses cryptography, distributed ledgers, peer-to-peer communications, scalable networks, and consensus. FlexiChain facilitates data structure agreements. This thesis offers a Block Directed Acyclic Graph (BDAG) architecture to link blocks to their forerunners to speed up validation. These data blocks are securely linked. This dissertation introduces Proof of Rapid Authentication, a novel consensus algorithm. This innovative method uses a distributed file to safely store a unique identifier (UID) based on node attributes to verify two blocks faster. This study also addresses CPS hardware security. A system of interconnected, user-unique identifiers allows each block's history to be monitored. This maintains each transaction and the validators who checked the block to ensure trustworthiness and honesty. We constructed a digital version that stays in sync with the distributed ledger as all nodes are linked by a NodeChain. The ledger is distributed without compromising node autonomy. Moreover, FlexiChain Layer 0 distributed ledger is also introduced and can connect and validate Layer 1 blockchains. This project produced a DAG-based blockchain integration platform with hardware security. The results illustrate a practical technique …