Emerging multicore Systems-on-Chip are enabling new cyber-physical applications such as autonomous drones, driverless cars and smart manufacturing using web-connected 3D printers. Common to those applications is a communicating task pipeline, to acquire and process sensor data and produce outputs that control actuators. As a result, these applications usually have timing requirements for both individual tasks and task pipelines formed for sensor data processing and actuation. Current cyber-physical programming platforms, such as Arduino and embedded Linux with the POSIX interface do not allow application developers to specify those timing requirements. Moreover, none of them provide the programming interface to schedule tasks and map them to processor cores, while managing I/O in a predictable manner, on multicore hardware platforms. Hence, this thesis presents the Qduino programming platform. Qduino adopts the simplicity of the Arduino API, with additional support for real-time multithreaded sketches on multicore architectures. Qduino allows application developers to specify timing properties of individual tasks as well as task pipelines at the design stage. To this end, we propose a mathematical framework to derive each task’s budget and period from the specified end-to-end timing requirements. The second part of the thesis is motivated by the observation that at the center of these pipelines are tasks that typically require complex software support, such as sensor data fusion or image processing algorithms. These features are usually developed by many man-year engineering efforts and thus commonly seen on General-Purpose Operating Systems (GPOS). Therefore, in order to support modern, intelligent cyber-physical applications, we enhance the Qduino platform’s extensibility by taking advantage of the Quest-V virtualized partitioning kernel. The platform’s usability is demonstrated by building a novel web-connected 3D printer and a prototypical autonomous drone framework in Qduino.

abstract: In recent years, deep learning systems have outperformed traditional machine learning systems in most domains. There has been a lot of research recently in the field of hand gesture recognition using wearable sensors due to the numerous advantages these systems have over vision-based ones. However, due to the lack of extensive datasets and the nature of the Inertial Measurement Unit (IMU) data, there are difficulties in applying deep learning techniques to them. Although many machine learning models have good accuracy, most of them assume that training data is available for every user while other works that do not require user data have lower accuracies. MirrorGen is a technique which uses wearable sensor data and generates synthetic videos using hand movements and it mitigates the traditional challenges of vision based recognition such as occlusion, lighting restrictions, lack of viewpoint variations, and environmental noise. In addition, MirrorGen allows for user-independent recognition involving minimal human effort during data collection. It also helps leverage the advances in vision-based recognition by using various techniques like optical flow extraction, 3D convolution. Projecting the orientation (IMU) information to a video helps in gaining position information of the hands. To validate these claims, we perform entropy analysis on various configurations such as raw data, stick model, hand model and real video. Human hand model is found to have an optimal entropy that helps in achieving user independent recognition. It also serves as a pervasive option as opposed to a video-based recognition. The average user independent recognition accuracy of 99.03% was achieved for a sign language dataset with 59 different users, 20 different signs with 20 repetitions each for a total of 23k training instances. Moreover, synthetic videos can be used to augment real videos to improve recognition accuracy. / Dissertation/Thesis / Masters Thesis Computer Science 2018

abstract: An examination of 12 darkweb sites involved in selling hacking services - often referred to as ”Hacking-as-a-Service” (HaaS) sites is performed. Data is gathered and analyzed for 7 months via weekly site crawling and parsing. In this empirical study, after examining over 200 forum threads, common categories of services available on HaaS sites are identified as well as their associated topics of conversation. Some of the most common hacking service categories in the HaaS market include Social Media, Database, and Phone hacking. These types of services are the most commonly advertised; found on over 50\% of all HaaS sites, while services related to Malware and Ransomware are advertised on less than 30\% of these sites. Additionally, an analysis is performed on prices of these services along with their volume of demand and comparisons made between the prices listed in posts seeking services with those sites selling services. It is observed that individuals looking to hire hackers for these services are offering to pay premium prices, on average, 73\% more than what the individual hackers are requesting on their own sites. Overall, this study provides insights into illicit markets for contact based hacking especially with regards to services such as social media hacking, email breaches, and website defacement. / Dissertation/Thesis / Masters Thesis Computer Science 2018

abstract: There currently exist various challenges in learning cybersecuirty knowledge, along with a shortage of experts in the related areas, while the demand for such talents keeps growing. Unlike other topics related to the computer system such as computer architecture and computer network, cybersecurity is a multidisciplinary topic involving scattered technologies, which yet remains blurry for its future direction. Constructing a knowledge graph (KG) in cybersecurity education is a first step to address the challenges and improve the academic learning efficiency. With the advancement of big data and Natural Language Processing (NLP) technologies, constructing large KGs and mining concepts, from unstructured text by using learning methodologies, become possible. The NLP-based KG with the semantic similarity between concepts has brought inspiration to different industrial applications, yet far from completeness in the domain expertise, including education in computer science related fields. In this research work, a KG in cybersecurity area has been constructed using machine-learning-based word embedding (i.e., mapping a word or phrase onto a vector of low dimensions) and hyperlink-based concept mining from the full dataset of words available using the latest Wikipedia dump. The different approaches in corpus training are compared and the performance based on different similarity tasks is evaluated. As a result, the best performance of trained word vectors has been applied, which is obtained by using Skip-Gram model of Word2Vec, to construct the needed KG. In order to improve the efficiency of knowledge learning, a web-based front-end is constructed to visualize the KG, which provides the convenience in browsing related materials and searching for cybersecurity-related concepts and independence relations. / Dissertation/Thesis / Masters Thesis Computer Science 2018

<p> Code clone detection tools find exact or similar pieces of code, known as code clones. Code clones are categorized into four types of increasing difficulty of detection, ranging from purely textual (Type I) to purely semantic (Type IV). Most clone detectors reported in the literature, work well up to Type III, which accounts for syntactic differences. In between Type III and Type IV, however, there lies a spectrum of clones that, although still exhibiting some syntactic similarities, are extremely hard to detect&mdash;the Twilight Zone. Besides correctness, scalability has become a must-have requirement for modern clone detection tools. The increase in amount of source code in web-hosted open source repository services has presented opportunities to improve the state of the art in various modern use cases of clone detection such as detecting similar mobile applications, license violation detection, mining library candidates, code repair, and code search among others. Though these opportunities are exciting, scaling such vast corpora poses critical challenge. </p><p> Over the years, many clone detection techniques and tools have been developed. One class of these techniques is based on software metrics. Metrics based clone detection has potential to identify clones in the Twilight Zone. For various reasons, however, metrics-based techniques are hard to scale to large datasets. My work highlights issues which prohibit metric based clone detection techniques to scale large datasets while maintaining high levels of correctness. The identification of these issues allowed me to rethink how metrics could be used for clone detection. </p><p> This dissertation starts by presenting an empirical study using software metrics to understand if metrics can be used to identify differences in cloned and non-cloned code. The study is followed by another large scale study to explore the extent of cloning in GitHub. Here, the dissertation highlights scalability challenges in clone detection and how they were addressed. The above two studies provided a strong base to use software metrics for clone detection in a scalable manner. To this end, the dissertation presents Oreo, a novel approach capable of detecting harder-to-detect clones in the Twilight Zone. Oreo is built using a combination of machine learning, information retrieval, and software metrics. This dissertation evaluates the recall of Oreo on BigCloneBench, a benchmark of real world code clones. In experiments to compare the detection performance of Oreo with other five state of the art clone detectors, we found that Oreo has both high recall and precision. More importantly, it pushes the boundary in detection of clones with moderate to weak syntactic similarity, in a scalable manner. Further, to address the issues identified in precision evaluations, the dissertation presents InspectorClone, a semi automated approach to facilitate precision studies of clone detection tools. InspectorClone makes use of some of the concepts introduced in the design of Oreo to automatically resolve different types of clone pairs. Experiments demonstrate that InspectorClone has a very high precision and it significantly reduces the number of clone pairs that need human validation during precision experiments. Moreover, InspectorClone aggregates the individual effort of multiple teams into a single evolving dataset of labeled clone pairs, creating an important asset for software clone research. Finally, the dissertation concludes with a discussion on the lessons learned during the design and development of Oreo and lists down a few areas for the future work in code clone detection.</p><p>

abstract: In recent years, the food, energy, and water (FEW) nexus has become a topic of considerable importance and has spurred research in many scientific and technical fields. This increased interest stems from the high level, and broad area, of impact that could occur in the long term if the interactions between these complex FEW sectors are incorrectly or only partially defined. For this reason, a significant amount of interdisciplinary collaboration is needed to accurately define these interactions and produce viable solutions to help sustain and secure resources within these sectors. Providing tools that effectively promote interdisciplinary collaboration would allow for the development of a better understanding of FEW nexus interactions, support FEW policy-making under uncertainty, facilitate identification of critical design requirements for FEW visualizations, and encourage proactive FEW visualization design. The goal of this research will be the completion of 3 primary objectives: (i) specify visualization design requirements relating to the FEW nexus; (ii) develop visualization approaches for the FEW nexus; and (iii) provide a comparison of current FEW visualization approaches against the proposed visualization approach. These objectives will be accomplished by reviewing graph-based visualization, network evolution, and visual analysis of volume data tasks, discussion with domain experts, examination of currently used visualization methods in FEW research, and conduction of a user study. This will provide a more thorough and representative depiction of the FEW nexus, as well as a basis for further research in the area of FEW visualization. This research will enhance collaboration between policymakers and domain experts in an attempt to encourage in-depth nexus research that will help support informed policy-making and promote future resource security. / Dissertation/Thesis / Masters Thesis Engineering 2019

abstract: Networks naturally appear in many high-impact applications. The simplest model of networks is single-layered networks, where the nodes are from the same domain and the links are of the same type. However, as the world is highly coupled, nodes from different application domains tend to be interdependent on each other, forming a more complex network model called multi-layered networks. Among the various aspects of network studies, network connectivity plays an important role in a myriad of applications. The diversified application areas have spurred numerous connectivity measures, each designed for some specific tasks. Although effective in their own fields, none of the connectivity measures is generally applicable to all the tasks. Moreover, existing connectivity measures are predominantly based on single-layered networks, with few attempts made on multi-layered networks. Most connectivity analyzing methods assume that the input network is static and accurate, which is not realistic in many applications. As real-world networks are evolving, their connectivity scores would vary by time as well, making it imperative to keep track of those changing parameters in a timely manner. Furthermore, as the observed links in the input network may be inaccurate due to noise and incomplete data sources, it is crucial to infer a more accurate network structure to better approximate its connectivity scores. The ultimate goal of connectivity studies is to optimize the connectivity scores via manipulating the network structures. For most complex measures, the hardness of the optimization problem still remains unknown. Meanwhile, current optimization methods are mainly ad-hoc solutions for specific types of connectivity measures on single-layered networks. No optimization framework has ever been proposed to tackle a wider range of connectivity measures on complex networks. In this thesis, an in-depth study of connectivity measures, inference, and optimization problems will be proposed. Specifically, a unified connectivity measure model will be introduced to unveil the commonality among existing connectivity measures. For the connectivity inference aspect, an effective network inference method and connectivity tracking framework will be described. Last, a generalized optimization framework will be built to address the connectivity minimization/maximization problems on both single-layered and multi-layered networks. / Dissertation/Thesis / Doctoral Dissertation Computer Science 2019

abstract: Capturing the information in an image into a natural language sentence is considered a difficult problem to be solved by computers. Image captioning involves not just detecting objects from images but understanding the interactions between the objects to be translated into relevant captions. So, expertise in the fields of computer vision paired with natural language processing are supposed to be crucial for this purpose. The sequence to sequence modelling strategy of deep neural networks is the traditional approach to generate a sequential list of words which are combined to represent the image. But these models suffer from the problem of high variance by not being able to generalize well on the training data. The main focus of this thesis is to reduce the variance factor which will help in generating better captions. To achieve this, Ensemble Learning techniques have been explored, which have the reputation of solving the high variance problem that occurs in machine learning algorithms. Three different ensemble techniques namely, k-fold ensemble, bootstrap aggregation ensemble and boosting ensemble have been evaluated in this thesis. For each of these techniques, three output combination approaches have been analyzed. Extensive experiments have been conducted on the Flickr8k dataset which has a collection of 8000 images and 5 different captions for every image. The bleu score performance metric, which is considered to be the standard for evaluating natural language processing (NLP) problems, is used to evaluate the predictions. Based on this metric, the analysis shows that ensemble learning performs significantly better and generates more meaningful captions compared to any of the individual models used. / Dissertation/Thesis / Masters Thesis Software Engineering 2019

abstract: There are many applications where the truth is unknown. The truth values are guessed by different sources. The values of different properties can be obtained from various sources. These will lead to the disagreement in sources. An important task is to obtain the truth from these sometimes contradictory sources. In the extension of computing the truth, the reliability of sources needs to be computed. There are models which compute the precision values. In those earlier models Banerjee et al. (2005) Dong and Naumann (2009) Kasneci et al. (2011) Li et al. (2012) Marian and Wu (2011) Zhao and Han (2012) Zhao et al. (2012), multiple properties are modeled individually. In one of the existing works, the heterogeneous properties are modeled in a joined way. In that work, the framework i.e. Conflict Resolution on Heterogeneous Data (CRH) framework is based on the single objective optimization. Due to the single objective optimization and non-convex optimization problem, only one local optimal solution is found. As this is a non-convex optimization problem, the optimal point depends upon the initial point. This single objective optimization problem is converted into a multi-objective optimization problem. Due to the multi-objective optimization problem, the Pareto optimal points are computed. In an extension of that, the single objective optimization problem is solved with numerous initial points. The above two approaches are used for finding the solution better than the solution obtained in the CRH with median as the initial point for the continuous variables and majority voting as the initial point for the categorical variables. In the experiments, the solution, coming from the CRH, lies in the Pareto optimal points of the multiobjective optimization and the solution coming from the CRH is the optimum solution in these experiments. / Dissertation/Thesis / Masters Thesis Computer Science 2019

Modern science and technology have witnessed in the past decade a proliferation of complex data that can be naturally modeled and interpreted as graphs. In real-world networked applications, the underlying graphs oftentimes exhibit fundamental community structures supporting widely varying interconnected processes. Identifying communities may offer insight on how the network is organized. In this thesis, we worked on community detection and search problems on graph data. Community detection (graph clustering) has become one of the most well-studied problems in graph management and analytics, the goal of which is to group vertices of a graph into densely knitted clusters with each cluster being well separated from all the others. Classic graph clustering methods primarily take advantage of topological information of graphs to model and quantify the proximity between vertices. With the proliferation of rich, heterogeneous graph contents widely available in real-world graphs, such as user profiles in social networks, it becomes essential to consider both structures and attributive contents of graphs for better quality graph clustering. On the other hand, existing community detection methods focus primarily on discovering communities in an apriori, top-down manner with the only reference to the input graph. As a result, all communities have to be exhaustively identified thus incurring expensive time/space cost and a huge amount of fruitless computation, if only a fraction of them are of special interest to end-users. In many real-world occasions, however, people are more interested in the communities pertaining to a given vertex. In our first project, we work on attributed graph clustering problem. We propose a graph embedding approach to cluster content-enriched, attributed graphs. The key idea is to design a unified latent representation for each vertex of a graph such that both the graph connectivity and vertex attribute proximity within the localized region of the vertex can be jointly embedded into a unified, continuous vector space. As a result, the challenging attributed graph clustering problem is cast to the traditional data clustering problem. In my second and third projects, we work on a query-dependent variant of community detection, referred to as the community search problem. The objective of community search is to identify dense subgraphs containing the query vertices. We study the community search problem in the truss-based model aimed at discovering all dense and cohesive k-truss communities to which the query set Q belongs. We introduce a novel equivalence relation, k-truss equivalence, to model the intrinsic density and cohesiveness of edges in k-truss communities and based on this equivalence we create 2 different space-efficient, truss-preserving index structure, EquiTruss and TEQ. Community search for one query or multiple queries can thus be addressed upon EquiTruss and TEQ without repeated, time-demanding accesses to the original graph, G, which proves to be theoretically optimal. While query set includes one query vertex in our first project, it includes multiple query vertices in our second project. As a summary, to get better quality on attributed graph clustering, the attribute-aware cluster information is well preserved during graph embedding. While we use Skip-Gram method for embedding, there are other embedding methods. We can use them to see the effect of different embedding methods on attributed graphs. In addition, our index structure is good for community search on large graphs without considering attribute information. Using attribute information in addition to the structure may give better communities for given query nodes. So, we can update our index structure to support community search on attributed graphs. / A Dissertation submitted to the Department of Computer Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2017. / November 6, 2017. / Community Detection, Community Search, Graph Embedding, Graph Mining, Indexing / Includes bibliographical references. / Peixiang Zhao, Professor Directing Dissertation; Washington Mio, University Representative; Piyush Kumar, Committee Member; Xiuwen Liu, Committee Member.