Computer Engineering Commons^™

Improving Facial Action Unit Recognition Using Convolutional Neural Networks, Shizhong Han

Theses and Dissertations

Recognizing facial action units (AUs) from spontaneous facial expression is a challenging problem, because of subtle facial appearance changes, free head movements, occlusions, and limited AU-coded training data. Most recently, convolutional neural networks (CNNs) have shown promise on facial AU recognition. However, CNNs are often overfitted and do not generalize well to unseen subject due to limited AU-coded training images. In order to improve the performance of facial AU recognition, we developed two novel CNN frameworks, by substituting the traditional decision layer and convolutional layer with the incremental boosting layer and adaptive convolutional layer respectively, to recognize the AUs from …

Improving Peptide Identification By Considering Ordered Amino Acid Usage, Ahmed Al-Qurri

Theses and Dissertations

Proteomics has made major progress in recent years after the sequencing of the genomes of a substantial number of organisms. A typical method for identifying peptides uses a database of peptides identified using tandem mass spectrometry (MS/MS). The profile of accurate mass and elution time (AMT) for peptides that need to be identified will be compared with this database. Restricting the search to those peptides detectable by MS will reduce processing time and more importantly increase accuracy. In addition, there are significant impacts for clinical studies. Proteotypic peptides are those peptides in a protein sequence that are most likely to …

A Sound Idea: An Investigation Into Accessible Video Game Design For The Deaf And Hard Of Hearing, Luke James Brook

Theses: Doctorates and Masters

A widely accepted, and incorrect, assumption towards hearing accessibility in video games is that deaf and hard of hearing (DHH) users are those who encounter the least barriers and are generally well catered for. Rapid advancement in video game technology has seen video game sound evolve from simple blips generated by internal circuitry to fully realised digital audio used to convey critical information. To accommodate the DHH, this information needs to be conveyed in an alternative manner. However, evidence suggests existing accessible design solutions for the DHH lack specificity and are insufficient. Thus, the inability to hear, or hear well, …

Development Of An Android Based Performance Assessment System For Motivational Interviewing Training, Sowmya Pappu

Browse all Theses and Dissertations

Motivational Interviewing (MI) has been proved to be an effective Screening, Brief Intervention, and Referral to Treatment (SBIRT) technique. It is an evidence-based practice used to identify, reduce, and prevent problematic use, abuse, and dependence on alcohol and illicit drugs. It emphasizes on patient-centered counseling approach that can help resolve their ambivalence through a non-confrontational, goal-oriented style for eliciting behavior change from the patient, almost like patients talk themselves into change. This approach provokes less resistance and stimulates the progress of patients at their own pace towards deciding about planning, making and sustaining positive behavioral change. Thus, training medical professionals …

Development Of An Ultra-Portable Non-Contact Wound Measurement System, Anka Babu Billa

Browse all Theses and Dissertations

Continuous monitoring of changes in wound size is key to correctly predict whether wounds will heal readily with conventional treatment or require more aggressive treatment strategies. Unfortunately, existing wound measurement solutions don't meet the clinical demand due to their limitations in accuracy, operating complexity and time, acquisition and operation cost, or reproducibility, resulting in unnecessarily lengthy recovery or extra treatment procedures, incurring an excessively high financial cost, and in many cases extended usage of addictive painkillers. In this thesis, we proposed and developed a low cost, a portable non-contact solution that combines multi-spectral imaging and a portfolio of imaging processing …

Efficient Reasoning Algorithms For Fragments Of Horn Description Logics, David Carral

Browse all Theses and Dissertations

We characterize two fragments of Horn Description Logics and we define two specialized reasoning algorithms that effectively solve the standard reasoning tasks over each of such fragments. We believe our work to be of general interest since (1) a rather large proportion of real-world Horn ontologies belong to some of these two fragments and (2) the implementations based on our reasoning approach significantly outperform state-of-the-art reasoners. Claims (1) and (2) are extensively proven via empirically evaluation.

Settings Protection Add-On: A User-Interactive Browser Extension To Prevent The Exploitation Of Preferences, Venkata Naga Siva Seelam

Browse all Theses and Dissertations

The abuse of browser preferences is a significant application security issue, despite numerous protections against automated software changing these preferences. Browser hijackers modify user’s desired preferences by injecting malicious software into the browser. Users are not aware of these modifications, and the unwanted changes can annoy the user and circumvent security preferences. Reverting these changes is not easy, and users often have to go through complicated sequences of steps to restore their preferences to the previous values. Tasks to resolve this issue include uninstalling and re-installing the browser, resetting browser preferences, and installing malware removal tools. This thesis describes a …

Distance Learning And Attribute Importance Analysis By Linear Regression On Idealized Distance Functions, Rupesh Kumar Singh

Browse all Theses and Dissertations

A good distance metric is instrumental on the performance of many tasks including classification and data retrieval. However, designing an optimal distance function is very challenging, especially when the data has high dimensions.Recently, a number of algorithms have been proposed to learn an optimal distance function in a supervised manner, using data with class labels. In this thesis we proposed methods to learn an optimal distance function that can also indicate the importance of attributes. Specifically, we present several ways to define idealized distance functions, two of which involving distance error correction involving KNN classification, and another involving a two-constant …

Semantics-Based Summarization Of Entities In Knowledge Graphs, Kalpa Gunaratna

Browse all Theses and Dissertations

The processing of structured and semi-structured content on the Web has been gaining attention with the rapid progress in the Linking Open Data project and the development of commercial knowledge graphs. Knowledge graphs capture domain-specific or encyclopedic knowledge in the form of a data layer and add rich and explicit semantics on top of the data layer to infer additional knowledge. The data layer of a knowledge graph represents entities and their descriptions. The semantic layer on top of the data layer is called the schema (ontology), where relationships of the entity descriptions, their classes, and the hierarchy of the …

An Optimization Compiler Framework Based On Polyhedron Model For Gpgpus, Lifeng Liu

Browse all Theses and Dissertations

General purpose GPU (GPGPU) is an effective many-core architecture that can yield high throughput for many scientific applications with thread-level parallelism. However, several challenges still limit further performance improvements and make GPU programming challenging for programmers who lack the knowledge of GPU hardware architecture. In this dissertation, we describe an Optimization Compiler Framework Based on Polyhedron Model for GPGPUs to bridge the speed gap between the GPU cores and the off-chip memory and improve the overall performance of the GPU systems. The optimization compiler framework includes a detailed data reuse analyzer based on the extended polyhedron model for GPU kernels, …

Oclep+: One-Class Intrusion Detection Using Length Of Patterns, Sai Kiran Pentukar

Browse all Theses and Dissertations

In an earlier paper, a method called One-class Classification using Length statistics of (jumping) Emerging Patterns (OCLEP) was introduced for masquerader detection. Jumping emerging patterns (JEPs) for a test instance are minimal patterns that match the test instance but they do not match any normal instances. OCLEP was based on the observation that one needs long JEPs to differentiate an instance of one class from instances of the same class, but needs short JEPs to differentiate an instance of one class from instances of a different class. In this thesis, we present OCLEP+, One-class Classification using Length statistics of Emerging …

Smart Ev Charging For Improved Sustainable Mobility, Ashutosh Shivakumar

Browse all Theses and Dissertations

The landscape of energy generation and utilization is witnessing an unprecedented change. We are at the threshold of a major shift in electricity generation from utilization of conventional sources of energy like coal to sustainable and renewable sources of energy like solar and wind. On the other hand, electricity consumption, especially in the field of transportation, due to advancements in the field of battery research and exponential technologies like vehicle telematics, is seeing a shift from carbon based to Lithium based fuel. Encouraged by 1. Decrease in the cost of Li – ion based batteries 2. Breakthroughs in battery chemistry …

Exploiting Alignments In Linked Data For Compression And Query Answering, Amit Krishna Joshi

Browse all Theses and Dissertations

Linked data has experienced accelerated growth in recent years due to its interlinking ability across disparate sources, made possible via machine-processable RDF data. Today, a large number of organizations, including governments and news providers, publish data in RDF format, inviting developers to build useful applications through reuse and integration of structured data. This has led to tremendous increase in the amount of RDF data on the web. Although the growth of RDF data can be viewed as a positive sign for semantic web initiatives, it causes performance bottlenecks for RDF data management systems that store and provide access to data. …

Conditional Correlation Analysis, Sanjeev Bhatta

Browse all Theses and Dissertations

Correlation analysis is a frequently used statistical measure to examine the relationship among variables in different practical applications. However, the traditional correlation analysis uses an overly simplistic method to do so. It measures how two variables are related in an application by examining only their relationship in the entire underlying data space. As a result, traditional correlation analysis may miss a strong correlation between those variables especially when that relationship exists in the small subpopulation of the larger data space. This is no longer acceptable and may lose a fair share of information in this era of Big Data which …

Intelligent Tutoring System Effects On The Learning Process, Ali Talib Qasim Al-Aqbi

Browse all Theses and Dissertations

The traditional education systems that have been used for several centuries have evolved very slowly and might be ineffective for addressing diverse learning styles and levels of preparation. This system is characterized by many students interacting with a single teacher, who is unable to address the individual needs of every student. Therefore, some students can become frustrated and fail to reach their educational potential. An Intelligent Tutoring System (ITS), which is a computer application used to provide students with one-to-one supplemental tutoring tailored to the student's learning style and pace, is of interest to educators for improving student learning. To …

Implementation And Evaluation Of Goal Selection In A Cognitive Architecture, Sravya Kondrakunta

Browse all Theses and Dissertations

A cognitive system attempts to achieve its goals by utilizing the appropriate resources present to yield the best possible outcome within a short duration. To achieve the goals in such an efficient manner, it is important for the agent to manage its goals well. Goal management not only makes the agent efficient but also flexible, more durable to the sudden changes in the environment, and self-reliant. Goal Management consists of various goal operations including goal formulation, selection, change, delegation, achievement, and monitoring. Each operation is unique and has its own significance in aiding the performance of the agent. The thesis …

Rendering Owl In Latex For Improved Readability: Extensions To The Owlapi, Cogan M. Shimizu

Browse all Theses and Dissertations

As ontology engineering is inherently a multidisciplinary process, it is necessary to utilize multiple vehicles to present an ontology to a user. In order to examine the content of an ontology, formal logic renderings of the axioms appear to be a very helpful approach for some. This thesis introduces a number of incremental improvements to the OWLAPI's \LaTeX{} rendering framework in order to improve the readability, concision, and correctness of OWL files translated into Description Logic and First Order Logic. In addition, we examine the efficacy of these renderings as vehicles for understanding an ontology.

Model Preparation And User Interface Aspects For Microsoft Hololens Medical Tutorial Applications, Andrew J. Mcnutt

Browse all Theses and Dissertations

Augmented Reality (AR) is the combination of computer vision and graphics to allow a device to derive context about the environment and to display relevant information to the user. The Microsoft Hololens is the latest wearable AR device that includes a series of infrared sensors to perform hand tracking and Simultaneous Localization and Mapping, a wave guide optics system to project holograms, and an independent untethered operating system. This thesis will explore the hardware and software capabilities of the device, prepare 3D graphical models, and understand the user interface. Specifically, in our research the target application is a tutorial on …

Diffusion Maps And Transfer Subspace Learning, Olga L. Mendoza-Schrock

Browse all Theses and Dissertations

Transfer Subspace Learning has recently gained popularity for its ability to perform cross-dataset and cross-domain object recognition. The ability to leverage existing data without the need for additional data collections is attractive for Aided Target Recognition applications. For Aided Target Recognition (or object assessment) applications, Transfer Subspace Learning is particularly useful, as it enables the incorporation of sparse and dynamically collected data into existing systems that utilize large databases. In this dissertation, Manifold Learning and Transfer Subspace Learning are combined to create new Aided Target Recognition systems capable of achieving high target recognition rates for cross-dataset conditions and cross-domain applications. …

Portable Motion Lab For Diagnostic And Rehabilitation Processes, Yogesh Laxman Chavan

Browse all Theses and Dissertations

In this thesis research, we have explored the capabilities of a household body-tracking device - Microsoft Xbox Kinect for motion analysis and assessment to assist patient’s rehabilitation after joint replacement and repair surgeries. In particular, we have combined the motion tracking capabilities of Kinect with the 3D modelling of human body for accurate and intuitive movement assessment and feedback. Our prototype system demonstrates the effective and efficient workflow of motion capturing, data processing and analyzing and augmented reality based presentation, resulting in a design which is not only accurate, reliable and robust, but also extremely cost- and space-efficient. Furthermore, we …

Visualization Of Carbon Monoxide Particles Released From Firearms, Sadan Suneesh Menon

Browse all Theses and Dissertations

A number of soldiers have come forward to report discomfort, irritation and respiratory problems after taking part in a live firing session. These problems are caused due to the fumes and particulates emitted from the gun upon firing. There exists substantial research work focused on lead and other harmful metallic particulates expelled from a firearm, since they are the most harmful among the other emissions. However, our research focuses on visualizing the carbon monoxide (CO) particles released from a firearm in order to help understand adverse effects they may have on the human body. We use data provided by researchers …

Web-Delivered Assembly Language Interactive Training And Its Sequence Identification For Software Reverse Engineering, Sunethra Thimmapuram

Browse all Theses and Dissertations

The purpose of this research is to help the beginners of software reverse engineering, the process of extracting knowledge or design information from a product and reproducing it on basis of extracted information. Beginner users often find it difficult to use, as the task may be found daunting at first sight due to a lack of existing learning resources. We need a better way to present information to the novice reverse engineer about how to understand low-level sequences of assembly instructions. This is akin to how the key to learning a foreign language is based upon a fundamental knowledge of …

Collaboratively Detecting Http-Based Distributed Denial Of Service Attack Using Software Defined Network, Ademola A. Ikusan

Browse all Theses and Dissertations

Over the years, there have been a lot of attacks on the web servers of critical companies and organization. As much as these organizations tighten their security, it always tends not to be enough to stop all the attacks towards their servers. These attack vectors are so difficult to stop because of the technicalities behind the attack. Furthermore, there are different classes of distributed denial of service(DDoS) attacks such as the high rates (SYN Flood attacks, ARP Flood attacks, etc), low-rate(HTTP-GET, HTTP-POST). Although there are different ways to stop high rate DDoS attacks by traffic sampling and detecting malformed packets …

Development Of A Performance Assessment System For Language Learning, Imen Kasrani

Browse all Theses and Dissertations

Recent advances in computer-assisted, language-speaking, learning/training technology have demonstrated its promising potential to improve the outcome of language learning in early education, special education, English as a Second Language (ESL), and foreign language. The growing number of readily available mobile app-based solutions help encourage interest in learning to speak a foreign language, but their effectiveness is limited due to their lack of objective assessment and performance feedback resembling expert judgment. For example, it has been recognized that, in early education, students learn best with one-on-one instructions. Unfortunately, teachers do not have the time, and it is challenging to extend the …

Finding Street Gang Member Profiles On Twitter, Lakshika Balasuriya

Browse all Theses and Dissertations

The crime and violence street gangs introduce into neighborhoods is a growing epidemic in cities around the world. Today, over 1.4 million people, belonging to more than 33,000 gangs, are active in the United States, of which 88% identify themselves as being members of a street gang. With the recent popularity of social media, street gang members have established online presences coinciding with their physical occupation of neighborhoods. Recent studies report that approximately 45% of gang members participate in online offending activities such as threatening, harassing individuals, posting violent videos or attacking someone on the street for something they said …

Adapting The Search Space While Limiting Damage During Learning In A Simulated Flapping Wing Micro Air Vehicle, Monica Sam

Browse all Theses and Dissertations

Cyber-Physical Systems (CPS) are characterized by closely coupled physical and software components that operate simultaneously on different spatial and temporal scales; exhibit multiple and distinct behavioral modalities; and interact with one another in ways not entirely predictable at the time of design. A commonly appearing type of CPS are systems that contain one or more smart components that adapt locally in response to global measurements of whole system performance. An example of a smart component robotic CPS system is a Flapping Wing Micro Air Vehicle (FW-MAV) that contains wing motion oscillators that control their wing flapping patterns to enable the …

Semantic Web Foundations For Representing, Reasoning, And Traversing Contextualized Knowledge Graphs, Vinh Thi Kim Nguyen

Browse all Theses and Dissertations

Semantic Web technologies such as RDF and OWL have become World Wide Web Consortium (W3C) standards for knowledge representation and reasoning. RDF triples about triples, or meta triples, form the basis for a contextualized knowledge graph. They represent the contextual information about individual triples such as the source, the occurring time or place, or the certainty. However, an efficient RDF representation for such meta-knowledge of triples remains a major limitation of the RDF data model. The existing reification approach allows such meta-knowledge of RDF triples to be expressed in RDF by using four triples per reified triple. While reification is …

Investigate Genomic 3d Structure Using Deep Neural Network, Yan Zhang

Theses and Dissertations

The 3D structures of the chromosomes play fundamental roles in essential cellular functions, e.g., gene regulation, gene expression, evolution and Hi-C technique provides the interaction density between loci on chromosomes. In this dissertation, we developed multiple algorithms, focusing the deep learning approach, to study the Hi-C datasets and the genomic 3D structures.

Building 3D structure of the genome one of the most critical purpose of the Hi-C technique. Recently, several approaches have been developed to reconstruct the 3D model of the chromosomes from HiC data. However, all of the methods are based on a particular mathematical model and lack of …

Blind Change Point Detection And Regime Segmentation Using Gaussian Process Regression, Sourav Das

Theses and Dissertations

Time-series analysis is used heavily in modeling and forecasting weather, economics, medical data as well as in various other fields. Change point detection (CPD) means finding abrupt changes in the time-series when the statistical property of a certain part of it starts to differ. CPD has attracted a lot of attention in the artificial intelligence, machine learning and data mining communities. In this thesis, a novel CPD algorithm is introduced for segmenting multivariate time-series data. The proposed algorithm is a general pipeline to process any high dimensional multivariate time-series data using nonlinear non-parametric dynamic system. It consists of manifold learning …