The amount of data and analysis being published and archived in the biomedical research community is more than can feasibly be sifted through manually, which limits the information an individual or small group can synthesize and integrate into their own research. This presents an opportunity for using automated methods, including Natural Language Processing (NLP), to extract important information from text on various topics. Named Entity Recognition (NER), is one way to automate knowledge extraction of raw text. NER is defined as the task of identifying named entities from text using labels such as people, dates, locations, diseases, and proteins. There are several NLP tools that are designed for entity recognition, but rely on large established corpus for training data. Biomedical research has the potential to guide diagnostic and therapeutic decisions, yet the overwhelming density of publications acts as a barrier to getting these results into a clinical setting. An exceptional example of this is the field of breast cancer biology where over 2 million people are diagnosed worldwide every year and billions of dollars are spent on research. Breast cancer biology literature and research relies on a highly specific domain with unique language and vocabulary, and therefore requires specialized NLP tools which can generate biologically meaningful results. This thesis presents a novel annotation tool, that is optimized for quickly creating training data for spaCy pipelines as well as exploring the viability of said data for analyzing papers with automated processing. Custom pipelines trained on these annotations are shown to be able to recognize custom entities at levels comparable to large corpus based recognition.

Lockout Tagout (LOTO) is a safety procedure instated by the Occupational Safety and Health Administration (OSHA) when doing maintenance on dangerous machinery and hazardous power sources. In this procedure, authorized workers shut off the machinery and use physical locks and tags to prevent operation during maintenance. LOTO has been the industry standard for 32 years since it was instantiated, being used in many different industries such as industrial work, mining, and agriculture. However, LOTO is not without its issues. The LOTO procedure requires employees to be trained and is prone to human error. As well, there is a clash between the technological advancement of machinery and the requirement of physical locks and tags required for LOTO. In this thesis, we propose a digital LOTO system to help streamline the LOTO procedure and increase the safety of the workers with machine learning. We first discuss what LOTO is, its current requirements, limitations, and issues. Then we look at current IoT locks and digital LOTO solutions and compare them to the requirements of traditional LOTO. Then we present our proposed digital LOTO system which will enhance the safety of workers and streamline the LOTO process with machine learning. Our digital LOTO system uses a rule-based system that enforces and streamlines the LOTO procedure and uses machine learning to detect potential violations of LOTO standards. We also validate that our system fulfills the requirements of LOTO and that the combination of machine learning and rule-based systems ensures the safety of workers by detecting violations with high accuracy. Finally, we discuss potential future work and improvements on this system as this thesis is part of a larger collaboration with Chevron, which plans to implement a digital LOTO system in their oil fields.

This report, "Analysis of System Reliability as a Capital Investment", is an analysis of radar system reliability of two similar tracking radar systems as a capital investment. It describes the two tracking radar systems and calculates the mission failures rates based upon field failure data. Additionally, an analysis of a simulation program written in FORTRAN is performed which treats system reliability as a capital investment based on 335 electronic systems that were fabricated with a reliability program versus 564 electronic systems fabricated without a reliability program. The data from the two tracking radar systems, one with reliability program and the other without, is incorporated in the computer program to verify the conclusions of the author of the computer simulation program.

Cloud computing increasingly handles confidential data, like private inference and query databases. Two strategies are used for secure computation: (1) employing CPU Trusted Execution Environments (TEEs) like AMD SEV, Intel SGX, or ARM TrustZone, and (2) utilizing emerging cryptographic methods like Fully Homomorphic Encryption (FHE) with libraries such as HElib, Microsoft SEAL, and PALISADE. To enhance computation, GPUs are often employed. However, using GPUs to accelerate secure computation introduces challenges addressed in three works. In the first work, we tackle GPU acceleration for secure computation with CPU TEEs. While TEEs perform computations on confidential data, extending their capabilities to GPUs is essential for leveraging their power. Existing approaches assume co-designed CPU-GPU setups, but we contend that co-designing CPU and GPU is difficult to achieve and requires early coordination between CPU and GPU manufacturers. To address this, we propose software-based memory encryption for CPU-GPU TEE co-design via the software layer. Yet, this introduces issues due to AES's 128-bit granularity. We present optimizations to mitigate these problems, resulting in execution time overheads of 1.1\% and 56\% for regular and irregular applications. In the second work, we focus on GPU acceleration for the CPU FHE library HElib, particularly for comparison operations on encrypted data. These operations are vital in Machine Learning, Image Processing, and Private Database Queries, yet their acceleration is often overlooked. We extend HElib to harness GPU acceleration for its resource-intensive components like BluesteinNTT, BluesteinFFT, and Element-wise Operations. Addressing memory separation, dynamic allocation, and parallelization challenges, we employ several optimizations to address these challenges. With all optimizations and hybrid CPU-GPU parallelism, we achieve a 11.1$\times$ average speedup over the state-of-the-art CPU FHE library. In our latest work, we concentrate on minimizing the ciphertext size by leveraging insights from algorithms, data access patterns, and application requirements to reduce the operational footprint of an FHE application, particularly targeting Neural Network inference tasks. Through the implementation of all three levels of ciphertext compression (precision reduction in comparisons, optimization of access patterns, and adjustments in data layout), we achieve a remarkable 5.6$\times$ speedup compared to the state-of-the-art GPU implementation in 100x\cite{100x}. Overcoming these challenges is crucial for achieving significant GPU-driven performance improvements. This dissertation provides solutions to these hurdles, aiming to facilitate GPU-based acceleration of confidential data computation.

<p>Service-Oriented Architecture (SOA) focuses on dividing the enterprise application layer of an enterprise system into components (as services) that have direct relationships with the business functionality of the enterprise. Web services, which are based on message exchanges, are the most widely adopted SOA technology. Web services provide web-accessible programs and devices that have been widely promoted for cloud computing environments. However, different types of web services are required to model actual services in the business domain. Particularly, enterprises (business providers such as banks, health care, and insurance companies) usually send their agents or other personnel (e.g., representatives, installers, maintainers, and trainers) to client sides to perform required services. An enterprise agent can be modeled as a software agent - a computer program that cannot be transmitted efficiently by communication messages. Lacking an efficient way to model the transmission of enterprise agents in traditional message based technologies restricts the application and usage of service-oriented architectures. The central problem addressed in this thesis is the need to develop an efficient SOA model for enterprise agents that will enable service providers to process client data locally at the client side.</p> <p>To address the research problem, the thesis proposes to model enterprise agents in SOA with a generic software agent called the Service Representative. This is a generic software agent which stays at the client side and can be customized by different service providers to process client data locally. Moreover, to employ a service representative, the thesis proposes a new type of web services called Task Services. While a traditional web service, called Data Service, processes client data completely at the server side, a task service is a web service with the capability of processing client data and resources partially or completely at the client side, using a Service Representative. Each task service assigns a task with three components to the generic service representative: task model, task knowledge, and task data. The task components are mapped to business components such as business process models, business rules and actions, and business data, where they can be efficiently transmitted by service messages.</p> <p>The combination of a service representative and task services provides an executable platform for service providers at the client side. Moreover, the client does not need to reveal its data, and hence privacy and security are maintained. Large volume client data is processed locally, causing less network traffic. Finally, real-time and event-triggered web services can be developed, based on the proposed approach.</p> <p>The main contributions and novelty of this research are: i) a domain independent computational model of enterprise agents in SOA to support a wide variety of client-processing tasks, ii) client- side web services which are compatible with typical server-side web services and comparable to other client-side processing technologies, iii) extensions of the SOA architecture by adding novel generic components including the service representative, the competition desk, and the service composition certifier, iv) provision of a formal model of client-side and server-side web services based on their construction of business components, v) empirical evaluations of the web service model in a number of different applications, using a prototype system, and vi) the application of the developed model to a number of target domains including the healthcare field. Furthermore, because client-side and server-side web services are complementary, a decision support model is provided that will assist service developers to decide upon the best service type for a web service.</p> / Doctor of Science (PhD)

The Internet of Things is a recent computing paradigm, de- fined by networks of highly connected things – sensors, actuators and smart objects – communicating across networks of homes, buildings, vehicles, and even people. The Internet of Things brings with it a host of new problems, from managing security on constrained devices to processing never before seen amounts of data. While cloud computing might be able to keep up with current data processing and computational demands, it is unclear whether it can be extended to the requirements brought forth by Internet of Things. Fog computing provides an architectural solution to address some of these problems by providing a layer of intermediary nodes within what is called an edge network, separating the local object networks and the Cloud. These edge nodes provide interoperability, real-time interaction, routing, and, if necessary, computational delegation to the Cloud. This paper attempts to evaluate Go, a distributed systems language developed by Google, in the context of requirements set forth by Fog computing. Similar methodologies of previous literature are simulated and benchmarked against in order to assess the viability of Go in the edge nodes of Fog computing architecture.

NAND Flash-based Solid State Disks have many attractive technical merits, such as low power consumption, light weight, shock resistance, sustainability of hotter operation regimes, and extraordinarily high performance for random read access, which makes SSDs immensely popular and be widely employed in different types of environments including portable devices, personal computers, large data centers, and distributed data systems. However, current SSDs still suffer from several critical inherent limitations, such as the inability of in-place-update, asymmetric read and write performance, slow garbage collection processes, limited endurance, and degraded write performance with the adoption of MLC and TLC techniques. To alleviate these limitations, we propose optimizations from both specific outside applications layer and SSDs' internal layer. Since SSDs are good compromise between the performance and price, so SSDs are widely deployed as second layer caches sitting between DRAMs and hard disks to boost the system performance. Due to the special properties of SSDs such as the internal garbage collection processes and limited lifetime, traditional cache devices like DRAM and SRAM based optimizations might not work consistently for SSD-based cache. Therefore, for the outside applications layer, our work focus on integrating the special properties of SSDs into the optimizations of SSD caches. Moreover, our work also involves the alleviation of the increased Flash write latency and ECC complexity due to the adoption of MLC and TLC technologies by analyzing the real work workloads.

There is a need for better predictive model that reduces the number of malicious URLs being sent through emails. This system should learn from existing metadata about URLs. The ideal solution for this problem would be able to learn from its predictions. For example, if it predicts a URL to be malicious, and that URL is deemed safe by the sandboxing environment, the predictor should refine its model to account for this data. The problem, then, is to construct a model with these characteristics that can make these predictions for the vast number of URLs being processed. Given that the current system does not employ machine learning methods, we intend to investigate multiple such models and summarize which of those might be worth pursuing on a large scale.

Text classification is used in information extraction and retrieval from a given text, and text classification has been considered as an important step to manage a vast number of records given in digital form that is far-reaching and expanding. This thesis addresses patent document classification problem into fifteen different categories or classes, where some classes overlap with other classes for practical reasons. For the development of the classification model using machine learning techniques, useful features have been extracted from the given documents. The features are used to classify patent document as well as to generate useful tag-words. The overall objective of this work is to systematize NASA’s patent management, by developing a set of automated tools that can assist NASA to manage and market its portfolio of intellectual properties (IP), and to enable easier discovery of relevant IP by users. We have identified an array of methods that can be applied such as k-Nearest Neighbors (kNN), two variations of the Support Vector Machine (SVM) algorithms, and two tree based classification algorithms: Random Forest and J48. The major research steps in this work consist of filtering techniques for variable selection, information gain and feature correlation analysis, and training and testing potential models using effective classifiers. Further, the obstacles associated with the imbalanced data were mitigated by adding synthetic data wherever appropriate, which resulted in a superior SVM classifier based model.

Today's cloud landscape has evolved computing infrastructure into a dynamic, high utilization, service-oriented paradigm. This shift has enabled the commoditization of large-scale storage and distributed computation, allowing engineers to tackle previously untenable problems without large upfront investment. A key enabler of flexibility in the cloud is the ability to transfer running virtual machines across subnets or even datacenters using live migration. However, live migration can be a costly process, one that has the potential to interfere with other applications not involved with the migration. This work investigates storage interference through experimentation with real-world systems and well-established benchmarks. In order to address migration interference in general, a buffering technique is presented that offloads the migration's read, eliminating interference in the majority of scenarios.