Computer Sciences Commons^™

Machine Learning: Several Advances In Linear Discriminant Analysis, Multi-View Regression And Support Vector Machine, Shuai Zheng

Computer Science and Engineering Dissertations - Archive

Machine learning technology is now widely used in engineering, science, finance, healthcare, etc. In this dissertation, we make several advances in machine learning technologies for high dimensional data analysis, image data classification, recommender systems and classification algorithms. In this big data era, many data are high dimensional data which is difficult to analyze. We propose two efficient Linear Discriminant Analysis (LDA) based methods to reduce data to low dimensions. Kernel alignment measures the degree of similarity between two kernels. We propose kernel alignment inspired LDA to find a subspace to maximize the alignment between subspace-transformed data kernel and class indicator …

Video-Based Face Recognition Using Deep Learning For Single Sample Per Person (Sspp) Surveillance Applications, Mostafa Parchami

Computer Science and Engineering Dissertations - Archive

Face Recognition (FR) is the task of identifying a person based on images of the face of the identity. Systems for video-based face recognition in video surveillance seek to recognize individuals of interest in real-time over a distributed network of surveillance cameras. These systems are exposed to challenging unconstrained environments, where the appearance of faces captured in videos varies according to pose, expression, illumination, occlusion, blur, scale, etc. In addition, facial models for matching must be designed using a single reference facial image per target individual captured from a high-quality still camera under controlled conditions. Deep learning has shown great …

Protecting Informations Assets Against Social Engineering Attack, Kushtrim Sadiku

Theses and Dissertations

Informacioni është monedha e re e bizneseve me qasje të menjëhershme nëpër kufijtë e korporatave dhe organizatave ndërkombëtare sepse pjesa më e madhe e botës tani është e lidhur nëpërmjet internetit. [1]

Përderisa kjo ka mundësuar lidhjen globale dhe rritjen e biznesit, gjithashtu ka nxitur edhe modifikime kinerbetike, vjedhje dhe shkatërrim.

Rritja e shpejtë dhe numri i sulmeve kundrejt kompanive do të thotë se pa programe adekuate dhe sisteme mbrojtëse ato janë jashtzakonisht të cenueshme.

Sulmet mbi asetet e informacionit, qoftë me qëllim të keq apo pa qëllim mund të dëmtoj seriozisht apo ta shkatrrojë një organizatë nëpërmjet dëmtimit të …

Automated Systems For Testing Android Applications To Detect Sensitive Information Leakage, Sarker Tanveer Ahmed Rumee

Computer Science and Engineering Dissertations - Archive

Smart phones have become an important daily companion and often used by users to store various private data such as contacts, photos, messages, various social network accounts etc. Users can furthermore extend the functionality of their phone by downloading applications (or apps) from various developers and online application stores. However, apps may misuse the data stored on the phone or obtained from the sensors and users do not have any direct means to track that. Hence, the need for improved mechanisms to better manage the privacy of user data is very important. There has been a lot of effort to …

Crowd Data Analytics And Optimization, Habibur Rahman

Computer Science and Engineering Dissertations - Archive

Crowdsourcing can be defined as outsourcing with crowd, where crowd refers to the online workers who are willing to complete simple tasks for small monetary compensation. The overwhelming reach of internet has enabled us to exploit crowd in an unprecedented way. Crowdsourcing, nowadays, is considered as a tool to solve both simple tasks (such as labeling ground truth, image recognition etc.) and complex tasks (such as collaborative writing, citizen journalism etc.). Furthermore, it is also used to solve computational problems such as Entity Resolution, Top-k, Group-by etc. While crowdsourcing provides us with plenty of opportunities, it also presents us with …

What Gets Media Attention And How Media Attention Evolves Over Time: Large-Scale Empirical Evidence From 196 Countries, Jisun An, Haewoon Kwak

Research Collection School Of Computing and Information Systems

It is known that news topics, covered more frequently and over longer periods of time, are considered to be important to the public. Hence, what gets media attention and how me- dia attention evolves over time has been studied for decades in communication study. However, previous studies are con- fined to a few countries or a few topics, mainly due to lack of longitudinal global data. In this work, we use a large-scale news data compiled from 196 countries to provide empirical analyses of media attention dynamics.

Hybridguard: A Principal-Based Permission And Fine-Grained Policy Enforcement Framework For Web-Based Mobile Applications, Phu Huu Phung, Abhinav Mohanty, Rahul Rachapalli, Meera Sridhar

Computer Science Faculty Publications

Web-based or hybrid mobile applications (apps) are widely used and supported by various modern hybrid app development frameworks. In this architecture, any JavaScript code, local or remote, can access available APIs, including JavaScript bridges provided by the hybrid framework, to access device resources. This JavaScript inclusion capability is dangerous, since there is no mechanism to determine the origin of the code to control access, and any JavaScript code running in the mobile app can access the device resources through the exposed APIs. Previous solutions are either limited to a particular platform (e.g., Android) or a specific hybrid framework (e.g., Cordova) …

Weighted Distributions: A Brief Review, Perspective And Characterizations, Aamir Saghir, Gholamhossein G. Hamedani, Sadaf Tazeem, Aneeqa Khadim

Mathematics, Statistics and Computer Science Faculty Research and Publications

The weighted distributions are widely used in many fields such as medicine, ecology and reliability, to name a few, for the development of proper statistical models. Weighted distributions are milestone for efficient modeling of statistical data and prediction when the standard distributions are not appropriate. A good deal of studies related to the weight distributions have been published in the literature. In this article, a brief review of these distributions is carried out. Implications of the differing weight models for future research as well as some possible strategies are discussed. Finally, characterizations of these distributions based on a simple relationship …

Plant Image Processing: 3d Volume Reconstruction, Hyperspectral Information Mining And Visualization, Shi Cao

School of Computing: Dissertations, Theses, and Student Research

Image processing techniques have been widely used in plant science for plant phenotyping studies. These fast algorithms are desired to process massive image data. In this thesis, we analyze RGB digital images taken from different view angles of plants and propose an efficient ad-hoc algorithm to identify structures of plants by 3D volume reconstruction techniques. We study hyperspectral images of plants and extend our scope to other images from different scientific disciplines. Obtaining the spectral and spatial information simul- taneously is a challenging task due to the high dimensionality of hyperspectral images. We first develop a real-time interactive tool for …

A Reinforcement Learning Approach To Autonomous Speed Control In Robotic Systems, Nima Aghli

Theses and Dissertations

Model-free reinforcement learning techniques have been successfully used in diverse robotic applications. In this thesis, we Implement the Q-learning algorithm as one of the most used model-free algorithms to find an optimal control signal for driving fast running trains on fixed tracks without flipping over or derailing. We examine the performance of the human driver and compare the results of reinforcement learning based controller to human driver performance. To test the proposed algorithm, a complete hardware and software testbed has been designed. We conclude that in simple tasks, human drivers perform identical to reinforcement learning algorithm, but in more complicated …

Machine Learning Techniques For The Development Of A Stratego Bot, Eric Joyce

Theses, Dissertations and Culminating Projects

Stratego is a two-player, non-stochastic, imperfect-information strategy game in which players try to locate and capture the opponent's flag. At the outset o f each game, players deploy their pieces in any arrangement they choose. Throughout play, each player knows the positions of the opponent’s pieces, but not the specific identities o f the opponent’s pieces. The game therefore involves deduction, bluffing, and a degree o f invention in addition to the sort o f planning familiar to perfect-information games like chess or backgammon.

Developing a strong A.l. player presents three major challenges. Firstly, a Stratego program must maintain states …

Cooperation Between Top-Down And Low-Level Markov Chains For Generating Rock Drumming, Chris Caulfield

Computer Science

Without heavy modification, the Markov chain is insufficient to handle the task of generating rock drum parts. This paper proposes a system for generating rock drumming that involves the cooperation between a top - down Markov chain that determi nes the structure of created drum parts and a low - level Markov chain that determines their contents. The goal of this system is to generate verse - or chorus - length drum parts that sound reminiscent of the drumming on its input pieces.

Effects Of An Impaired Sonic Hedgehog Signaling Pathway And A Nonfunctional Gli3 Protein On Gnrh-1 Neuronal Migration In Gli3xt/Xt Mutants, Elizabet Aleks Genis

Computer Science

Gonadotropin releasing hormone (GnRH) is the master regulatory hormone for sexual development. During embryonic development, gonadotropin releasing hormone-1 neurons (GnRH-1ns) form in the olfactory pit and migrate, along axonal Peripherin positive fibers, from the nasal area to the pre-optic area of the basal forebrain. Upon migration into the brain, GnRH-1ns release GnRH. Defective migration of GnRH-1ns can result in hypogonadotropic hypogonadism (HH), a condition that results in lack of sexual development and infertility. When HH appears associated with reduced or absent sense of smell, it is clinically defined as Kallmann Syndrome (KS) (Paolo E Forni & Wray, 2015). The neurons …

High-Fidelity Spectrum Characterization With Low-Cost Sensors, Stuti Misra

Computer Science

With the increasing use of wireless technologies, we see a heavy use of the spectrum at certain frequencies whereas it is underutilized at other frequencies. We need to utilize the currently underutilized spectrum. Hence, a paradigm called Dynamic Spectrum Access arises. Dynamic Spectrum Access looks for opportunity to utilize this underutilized spectrum by allowing devices to opportunistically access spectrum that is not actively used. DSA, however, requires spectrum sensing and spectrum characterization across time, space, and frequency for opportunistic devices to know where to operate. Spectrum sensing is the process of collecting power level traces from the radio-frequency spectrum, whereas …

Inside The Mind Of Mcmillen, Tina Duong

ART 108: Introduction to Games Studies

The paper discusses how Edmund McMillen created “The Binding of Isaac”. It goes into his inspirations, influences, and his background leading up to the game. It then describes the game, going into its mechanics, design, art and story and how each of these elements were unique and risky to put into an Indie game. I discussed how “The Binding of Isaac” brought in fresh, new elements that most mainstream games wouldn’t dare touch, and despite that “Binding of Isaac” accomplished amazing success. Then, I talk about how the game was initially made by a group of two people, and was …

Mining Helpdesk Databases For Professional Development Topic Discovery, Joel T. Lowsky

All Theses And Dissertations

This single-site, instrumental case study created and tested a methodological road map by which academic institutions can use text data mining techniques to derive technology skillset weaknesses and professional development topics from the site’s technical support helpdesk database. The methods employed were described in detail and applied to the helpdesk database of an independent, co-educational boarding high school in the northeastern United States. Standard text data mining procedures, including the formation of a wordlist (frequently occurring terms), and the creation and application of clustering (automated data grouping) and classification (automated data labeling) models generated meaningful and revealing themes from the …

Exploiting Semantic Distance In Linked Open Data For Recommendation, Sultan Dawood Alfarhood

Graduate Theses and Dissertations

The use of Linked Open Data (LOD) has been explored in recommender systems in different ways, primarily through its graphical representation. The graph structure of LOD is utilized to measure inter-resource relatedness via their semantic distance in the graph. The intuition behind this approach is that the more connected resources are to each other, the more related they are. One drawback of this approach is that it treats all inter-resource connections identically rather than prioritizing links that may be more important in semantic relatedness calculations. Another drawback of current approaches is that they only consider resources that are connected directly …

A Hybrid Partially Reconfigurable Overlay Supporting Just-In-Time Assembly Of Custom Accelerators On Fpgas, Zeyad Tariq Aklah

Graduate Theses and Dissertations

The state of the art in design and development flows for FPGAs are not sufficiently mature to allow programmers to implement their applications through traditional software development flows. The stipulation of synthesis as well as the requirement of background knowledge on the FPGAs' low-level physical hardware structure are major challenges that prevent programmers from using FPGAs. The reconfigurable computing community is seeking solutions to raise the level of design abstraction at which programmers must operate, and move the synthesis process out of the programmers' path through the use of overlays. A recent approach, Just-In-Time Assembly (JITA), was proposed that enables …

Hierarchical Active Learning Application To Mitochondrial Disease Protein Dataset, James D. Duin

School of Computing: Dissertations, Theses, and Student Research

This study investigates an application of active machine learning to a protein dataset developed to identify the source of mutations which give rise to mitochondrial disease. The dataset is labeled according to the protein's location of origin in the cell; whether in the mitochondria or not, or a specific target location in the mitochondria's outer or inner membrane, its matrix, or its ribosomes. This dataset forms a labeling hierarchy. A new machine learning approach is investigated to learn the high-level classifier, i.e., whether the protein is a mitochondrion, by separately learning finer-grained target compartment concepts and combining the results. This …

Improving Software Quality By Synergizing Effective Code Inspection And Regression Testing, Bo Guo

Student Work

Software quality assurance is an essential practice in software development and maintenance. Evolving software systems consistently and safely is challenging. All changes to a system must be comprehensively tested and inspected to gain confidence that the modified system behaves as intended. To detect software defects, developers often conduct quality assurance activities, such as regression testing and code review, after implementing or changing required functionalities. They commonly evaluate a program based on two complementary techniques: dynamic program analysis and static program analysis. Using an automated testing framework, developers typically discover program faults by observing program execution with test cases that encode …

Applications Of Graph Embedding In Mesh Untangling, Jake Quinn

Student Work

The subject of this thesis is mesh untangling through graph embedding, a method of laying out graphs on a planar surface, using an algorithm based on the work of Fruchterman and Reingold[1]. Meshes are a variety of graph used to represent surfaces with a wide number of applications, particularly in simulation and modelling. In the process of simulation, simulated forces can tangle the mesh through deformation and stress. The goal of this thesis was to create a tool to untangle structured meshes of complicated shapes and surfaces, including meshes with holes or concave sides. The goals of graph embedding, such …

Molecular Dynamics Simulations Of Dna-Functionalized Nanoparticle Building Blocks On Gpus, Tyler Landon Fochtman

Graduate Theses and Dissertations

This thesis discusses massively parallel molecular dynamics simulations of nBLOCKs using graphical processing units. nBLOCKs are nanoscale building blocks composed of gold nanoparticles functionalized with single-stranded DNA molecules. To explore greater simulation time scales we implement our nBLOCK computational model as an extension to the coarse grain molecular simulator oxDNA. oxDNA is parameterized to match the thermodynamics of DNA strand hybridization as well as the mechanics of single stranded DNA and double stranded DNA. In addition to an in-depth review of our implementation details we also provide results of the model validation and performance tests. These validation and performance tests …

Characterizing And Improving Power And Performance In Hpc Networks, Taylor L. Groves

Computer Science ETDs

Networks are the backbone of modern HPC systems. They serve as a critical piece of infrastructure, tying together applications, analytics, storage and visualization. Despite this importance, we have not fully explored how evolving communication paradigms and network design will impact scientific workloads. As networks expand in the race towards Exascale (1×10^18 floating point operations a second), we need to reexamine this relationship so that the HPC community better understands (1) characteristics and trends in HPC communication; (2) how to best design HPC networks to save power or enhance the performance; (3) how to facilitate scalable, informed, and dynamic decisions within …

Lighttouch: Securely Connecting Wearables To Ambient Displays With User Intent, Xiaohui Liang, Tianlong Yun, Ronald Peterson, David Kotz

Dartmouth Scholarship

Wearables are small and have limited user interfaces, so they often wirelessly interface with a personal smartphone/computer to relay information from the wearable for display or other interactions. In this paper, we envision a new method, LightTouch, by which a wearable can establish a secure connection to an ambient display, such as a television or a computer monitor, while ensuring the user's intention to connect to the display. LightTouch uses standard RF methods (like Bluetooth) for communicating the data to display, securely bootstrapped via the visible-light communication (the brightness channel) from the display to the low-cost, low-power, ambient light sensor …

Using A Multi Variate Pattern Analysis (Mvpa) Approach To Decode Fmri Responses To Fear And Anxiety., Sajjad Torabian Esfahani

Electronic Theses and Dissertations

This study analyzed fMRI responses to fear and anxiety using a Multi Variate Pattern Analysis (MVPA) approach. Compared to conventional univariate methods which only represent regions of activation, MVPA provides us with more detailed patterns of voxels. We successfully found different patterns for fear and anxiety through separate classification attempts in each subject’s representational space. Further, we transformed all the individual models into a standard space to do group analysis. Results showed that subjects share a more common fear response. Also, the amygdala and hippocampus areas are more important for differentiating fear than anxiety.

Peeking Into The Other Half Of The Glass : Handling Polarization In Recommender Systems., Mahsa Badami

Electronic Theses and Dissertations

This dissertation is about filtering and discovering information online while using recommender systems. In the first part of our research, we study the phenomenon of polarization and its impact on filtering and discovering information. Polarization is a social phenomenon, with serious consequences, in real-life, particularly on social media. Thus it is important to understand how machine learning algorithms, especially recommender systems, behave in polarized environments. We study polarization within the context of the users' interactions with a space of items and how this affects recommender systems. We first formalize the concept of polarization based on item ratings and then relate …

Integrating Virtual Reality With Use-Of-Force Training Simulations, Ted Mader

Senior Honors Theses

No abstract provided.

Storing And Rendering Geospatial Data In Mobile Applications, Samip Neupane

Senior Honors Theses

Geographical Information Systems and geospatial data are seeing widespread use in various internet and mobile mapping applications. One of the areas where such technologies can be particularly valuable is aeronautical navigation. Pilots use paper charts for navigation, which, in contrast to modern mapping software, have some limitations. This project aims to develop an iOS application for phones and tablets that uses a GeoPackage database containing aeronautical geospatial data, which is rendered on a map to create an offline, feature-based mapping software to be used for navigation. Map features are selected from the database using R-Tree spatial indices. The attributes from …

Eit Imaging Of Admittivities With A D-Bar Method And Spatial Prior: Experimental Results For Absolute And Difference Imaging, Sarah J. Hamilton

Mathematics, Statistics and Computer Science Faculty Research and Publications

Electrical impedance tomography (EIT) is an emerging imaging modality that uses harmless electrical measurements taken on electrodes at a body's surface to recover information about the internal electrical conductivity and or permittivity. The image reconstruction task of EIT is a highly nonlinear inverse problem that is sensitive to noise and modeling errors making the image reconstruction task challenging. D-bar methods solve the nonlinear problem directly, bypassing the need for detailed and time-intensive forward models, to provide absolute (static) as well as time-difference EIT images. Coupling the D-bar methodology with the inclusion of high confidence a priori data results in a …

Diversity And Efficiency: An Unexpected Result, Joseph Smith Johnson

Theses and Dissertations

Empirical evidence shows that ensembles with adequate levels of pairwise diversity among a set of accurate member algorithms significantly outperform any of the individual algorithms. As a result, several diversity measures have been developed for use in optimizing ensembles. We show that diversity measures that properly combine the diversity space in an additive and multiplicative manner, not only result in ensembles whose accuracy is comparable to the naive ensemble of choosing the most accurate learners, but also results in ensembles that are significantly more efficient than such naive ensembles. In addition to diversity measures found in the literature, we submit …