Computer Sciences Commons^™

Analizmi Dhe Vizualizimi I Të Dhënave Për Lindjet Dhe Vdekjet Në Kosovë, Erijona Gashi

Theses and Dissertations

Në ditët e sotme është bërë pjesë e pashmangshme e çdo institucioni dhe biznesi trajtimi i të dhënave e sidomos i atyre të ndjeshme ku kërkohet kujdes më i shtuar. Përveç të dhënave që janë të nevojshme për të rritur dhe planifikuar resurset e kompanive, sot flasim edhe për një koncept tjetër, atë të të dhënave personale që ruajnë integritetin personal dhe financiar të personave të punësuar apo që shërbehen nga bizneset. Sot sa herë që shikohet suksesi i një kompanie, përveç produkteve cilësore që i ofron kompania shikohet edhe trajtimi që i bëhet të dhënave personale, pse ruhen ato, …

Shërbimet Dhe Kontrollet E Sigurisë Ne Sigurin E Sistemeve Të Informacionit, Besijana Fejzullahu

Theses and Dissertations

Kompani apo biznese të ndryshme për një periudhe shumë të gjatë janë munduar të mbledhin njohuri dhe të dhëna rreth shërbimëve ne sigurin e informacioneve,mirpo shumë pak nga këto biznise jan në njohuri rreth kësaj fushe. Prandaj,qëllimi i këtij studmi është që të jepet një kontribut shkencor rreth kërcënimëve apo dobësive të reja,kontrollat në fushën e informacionit,si dhe njohurit se si të zhvillohet me më efikasitet vlersimet dhe pamja programatike e rrezikut. Do ti vëm në pahë disa këshilla se si të reflektojmë dhe të bëhemi të suksesshëm në treg,ku nga këto benifite shumë kompani apo biznese mund të zhvillohen …

Mbledhja Dhe Analiza E Të Dhënave Dhe Aktiviteteve Nga Rrjetet Sociale Duke Përdorur Platforma Të Hapura, Afrim Rexhepi

Theses and Dissertations

Në këtë punim është paraqitur dhe diskutuar i gjithë procesi i konfigurimit, mbledhjes, ruajtjes dhe analizës së shënimeve nga rrjeti social Twitter, duke përdorur të gjitha platformat e hapura (Open Source) - pa pagesë.

Punimi përfshinë pjesën teorike dhe pjesën praktike apo realizimin teknik të procesit për mbledhjen, ruajtjen dhe analizën e të dhënave (Data Science). Në pjesën teorike është diskutuar për të gjeturat dhe analizat nga publikime dhe artikuj të ndryshëm lidhur me të dhënat nga rrjetet sociale. Gjithashtu janë elaboruar edhe metodat e ndryshme të analizës së përmbajtjes së tekstit. Në pjesën e realizimit teknik është paraqitur në …

Menaxhimi I Shkollës Private Parauniversitare Përmes Vdi, Zamire Ramadani

Theses and Dissertations

Me rritjen e numrit të kompjuterëve personal të vendosur në shkolla, ndërmarrje dhe organizata të ndryshme, rritet edhe ngarkesa dhe përgjegjësia e administratorëve të sistemit për shkak problemeve që lidhen me konsumin e energjisë, shpenzimet e IT, shpenzimet e zëvendësimit të PC, kapacitetin e ruajtjes së të dhënave dhe sigurinë e informacionit. Sidoqoftë, është vërtetuar se aplikimi i virtualizimit është një zgjidhje e suksesshme e cila ka kosto efektive për t’i zgjidhur këto probleme.

Virtualizimi është teknologjia kryesore e cila ofron zgjidhje efikase të IT në sektorët e arsimit dhe ndërmarrjeve. Ne paraqesim një kornizë origjinale strukturore për të cilën …

Interneti I Gjërave Dhe Impakti I Tij Në Privatësinë E Të Dhënave, Fakete Mustafa

Theses and Dissertations

Kjo tezë përpiqet të kuptojë mardhënjet e Internet of Things (IoT) me privatësin dhe Vetëdijsimin. Teza përdor një metodë sasiore për të kuptuar këto marrëdhënie nëpërmjet lidhjes së korrelacionit dhe regresionit linear. Paketa statistikore për Shkencat Sociale (SPSS) është përdorur për të përpunuar të dhënat e mbledhura përmes pyetësorëve. Sipas analizës, ekziston një korrelacion i fortë midis Internet of Things (IoT) dhe privatësis. Për më tepër, korrelacioni midis Internet of Things (IoT) dhe Vetëdijsimin është gjithashtu pozitiv. Analiza e regresionit linear tregon që privatësia është shume e rëndësishëme në përdorimin e paisjeve Internet of Things, ndërsa vetëdijsimi eliminohet nga ekuacioni …

Krahasimi I Javascript Librarive – React Js Me Vue Js, Ylber Verbaj

Theses and Dissertations

Implementimi i teklogjive moderne në zhvillimin e web-it po merr popullaritet global dhe po rezulton të jetë tejet i suksesshëm në këtë industri. Rezultatet e mira dalin të jenë në rrafshin professional dhe atë personal të zhvilluesve. Rritja dhe zhvillimi i këtyre teknologjive në mënyrë eksponenciale ka shtyrë shume kompani dhe shumë zhvillues se cila prej këtyre teknologjive gjen zbatim të duhur në produktet përkatëse të tyre.

Në këtë hulumtim do të shtjelloj elementet e web-it, mënyrën e funksionimit, ndarjet e web-it në anën e logjikës(backend) dhe ndërfaqës së përdoruesit(frontend). Hulumtimi kryesorë është në pjesën e frontend ku krahasohen dy …

- Aplikimi I Njohjes Së Fytyrës Duke Shfrytëzuar Shërbimet-Aws, Fatson Sylejmani

Theses and Dissertations

Sistemi i njohjes së fytyrës është njëri nga proceset kryesore biometrike të informacionit pasi ka gjetë zbatim shumë të madh në industri të ndryshme. Sistemi i tillë është mjaft kompleks por edhe shumë më efikas dhe më i besueshëm krahasuar me proceset tjera biometrike si: skanimi i irisit, nënshkrimi, gjurmët e gishtave.

Zhvillimi i një sistemi të tillë për njohjen e fytyrës i cili mund të përdoret për arsye të ndryshme ka qenë dhe është në interes për shumë zhvillues dhe kërkues në dy dekadat e fundit. Disa nga arsyet kryesore janë nevoja e njohjeve automatike dhe sistemet e mbikqyrjes, …

Dissemination And Visualization Of Hydro-Climate Data In Sub-Saharan Africa For Analysis Of Climatic Parameters, Divyadharshini Karthikeyan, Aparna S. Varde, Clement Alo

School of Computing Faculty Scholarship and Creative Works

Precipitation can have adverse effects on the climate ecosystem. Too much can impose concerns such as flooding and landslides, resulting in damaged property, agricultural losses, and loss of life. Too little, and drought becomes an issue, inducing wildfires, poor air quality, agricultural losses, and health degradation. While much work has been performed on historical and projected analysis of heavy precipitation, few interactive visualizations exist for end-users to better understand local impacts. The goal of this project is to create a visualization tool that easily demonstrates how precipitation extremes have changed and might change in the future for Sub-Saharan Africa. This …

Bots And Humans On Social Media, Lale Madahali

Interdisciplinary Informatics Faculty Proceedings & Presentations

Social networks are an important part of today’s life. They are used for entertainment, getting the news, advertisements, and branding for businesses and individuals alike. Research shows that automated accounts, also known as bots, contribute to the content spread on social media allowing the the environment pollution and public opinion manipulation. This research aims at investigating bots’ behavior on Twitter and examine how different and similar they are compared to humans. I will investigate their underlying network, whether it is an information network or social network. In the second step, I attempt to answer whether they follow the structure of …

Set Operators, Xiaojin Ye

Dissertations, Theses, and Capstone Projects

My research is centered on set operators. These are universally applicable regardless of the internal structure (numeric or non-numeric) of each individual observed datum. In our research, we have developed the theory of set operators to fill holes and gaps in observed data and eliminate paper shred garbage, thereby changing the observed symbolic data set into one whose pattern is closer to the pattern in the underlying population from which the observed data set was sampled with perturbations.

We describe different set operators including increasing operators, decreasing operators, ex- pansive operators, contractive operators, union preserving operators, intersection preserving op- erators, …

Marble: Model-Based Robustness Analysis Of Stateful Deep Learning Systems, Xiaoning Du, Yi Li, Xiaofei Xie, Lei Ma, Yang Liu, Jianjun Zhao

Research Collection School Of Computing and Information Systems

State-of-the-art deep learning (DL) systems are vulnerable to adversarial examples, which hinders their potential adoption in safetyand security-critical scenarios. While some recent progress has been made in analyzing the robustness of feed-forward neural networks, the robustness analysis for stateful DL systems, such as recurrent neural networks (RNNs), still remains largely uncharted. In this paper, we propose Marble, a model-based approach for quantitative robustness analysis of real-world RNN-based DL systems. Marble builds a probabilistic model to compactly characterize the robustness of RNNs through abstraction. Furthermore, we propose an iterative refinement algorithm to derive a precise abstraction, which enables accurate quantification of …

Weakly Paired Multi-Domain Image Translation, M.Y. Zhang, Zhiwu Huang, D.P. Paudel, J. Thoma, Gool L. Van

Research Collection School Of Computing and Information Systems

In this paper, we aim at studying the new problem of weakly paired multi-domain image translation. To this end, we collect a dataset that contains weakly paired images from multiple domains. Two images are considered to be weakly paired if they are captured from nearby locations and share an overlapping field of view. These images are possibly captured by two asynchronous cameras—often resulting in images from separate domains, e.g. summer and winter. Major motivations for using weakly paired images are: (i) performance improvement towards that of paired data; (ii) cheap labels and abundant data availability. For the first time in …

Hierarchical Multimodal Attention For End-To-End Audio-Visual Scene-Aware Dialogue Response Generation, Hung Le, Doyen Sahoo, Nancy F. Chen, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

This work is extended from our participation in the Dialogue System Technology Challenge (DSTC7), where we participated in the Audio Visual Scene-aware Dialogue System (AVSD) track. The AVSD track evaluates how dialogue systems understand video scenes and responds to users about the video visual and audio content. We propose a hierarchical attention approach on user queries, video caption, audio and visual features that contribute to improved evaluation results. We also apply a nonlinear feature fusion approach to combine the visual and audio features for better knowledge representation. Our proposed model shows superior performance in terms of both objective evaluation and …

Pricing And Equilibrium In On-Demand Ride-Pooling Markets, Jintao Ke, Hai Yang, Xinwei Li, Hai Wang, Jieping Ye

Research Collection School Of Computing and Information Systems

With the recent rapid growth of technology-enabled mobility services, ride-sourcing platforms, such as Uber and DiDi, have launched commercial on-demand ride-pooling programs that allow drivers to serve more than one passenger request in each ride. Without requiring the prearrangement of trip schedules, these programs match on-demand passenger requests with vehicles that have vacant seats. Ride-pooling programs are expected to offer benefits for both individual passengers in the form of cost savings and for society in the form of traffic alleviation and emission reduction. In addition to some exogenous variables and environments for ride-sourcing market, such as city size and population …

How (Not) To Find Bugs: The Interplay Between Merge Conflicts, Co-Changes, And Bugs, Luis Amaral, Marcos C. Oliveira, Welder Luz, José Fortes, Rodrigo Bonifacio, Daniel Alencar, Eduardo Monteiro, Gustavo Pinto, David Lo

Research Collection School Of Computing and Information Systems

Context: In a seminal work, Ball et al. [1] investigate if the information available in version control systems could be used to predict defect density, arguing that practitioners and researchers could better understand errors "if [our] version control system could talk". In the meanwhile, several research works have reported that conflict merge resolution is a time consuming and error-prone task, while other contributions diverge about the correlation between co-change dependencies and defect density. Problem: The correlation between conflicting merge scenarios and bugs has not been addressed before, whilst the correlation between co-change dependencies and bug density has been only investigated …

Deepstyle: User Style Embedding For Authorship Attribution Of Short Texts, Zhiqiang Hu, Roy Ka-Wei Lee, Lei Wang, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Authorship attribution (AA), which is the task of finding the owner of a given text, is an important and widely studied research topic with many applications. Recent works have shown that deep learning methods could achieve significant accuracy improvement for the AA task. Nevertheless, most of these proposed methods represent user posts using a single type of features (e.g., word bi-grams) and adopt a text classification approach to address the task. Furthermore, these methods offer very limited explainability of the AA results. In this paper, we address these limitations by proposing DeepStyle, a novel embedding-based framework that learns the representations …

Optimal Control Of Excitable Systems Near Criticality, Kathleen Finlinson, Woodrow L. Shew, Danile B. Larremore, Juan G. Restrepo

Physics Faculty Publications and Presentations

Experiments suggest that the cerebral cortex gains several functional advantages by operating in a dynamical regime near the critical point of a phase transition. However, a long-standing criticism of this hypothesis is that critical dynamics are rather noisy, which might be detrimental to aspects of brain function that require precision. If the cortex does operate near criticality, how might it mitigate the noisy fluctuations? One possibility is that other parts of the brain may act to control the fluctuations and reduce cortical noise. To better understand basic aspects of controlling neural activity fluctuations, here we numerically and analytically study a …

Cats Are Not Fish: Deep Learning Testing Calls For Out-Of-Distribution Awareness, David Berend, Xiaofei Xie, Lei Ma, Lingjun Zhou, Yang Liu, Chi Xu, Jianjun Zhao

Research Collection School Of Computing and Information Systems

As Deep Learning (DL) is continuously adopted in many industrial applications, its quality and reliability start to raise concerns. Similar to the traditional software development process, testing the DL software to uncover its defects at an early stage is an effective way to reduce risks after deployment. According to the fundamental assumption of deep learning, the DL software does not provide statistical guarantee and has limited capability in handling data that falls outside of its learned distribution, i.e., out-of-distribution (OOD) data. Although recent progress has been made in designing novel testing techniques for DL software, which can detect thousands of …

Querying Recurrent Convoys Over Trajectory Data, Munkh-Erdene Yadamjav, Zhifeng Bao, Baihua Zheng, Farhana M. Choudhury, Hanan Samet

Research Collection School Of Computing and Information Systems

Moving objects equipped with location-positioning devices continuously generate a large amount of spatio-temporal trajectory data. An interesting finding over a trajectory stream is a group of objects that are travelling together for a certain period of time. Existing studies on mining co-moving objects do not consider an important correlation between co-moving objects, which is the reoccurrence of the movement pattern. In this study, we define a problem of finding recurrent pattern of co-moving objects from streaming trajectories and propose an efficient solution that enables us to discover recent co-moving object patterns repeated within a given time period. Experimental results on …

Design And Implementation Of A Secure Access Layer For A Psd2 Compliant Consent Management Engine, Mentor Elshani

Theses and Dissertations

Online banking is growing fast, and customers are becoming increasing more comfortable with it than with the traditional bank services. Banks have to remain up to date with the latest trends in technology to satisfy customers’ demands for their everyday usage of banking services. Various banks often offer mobile applications with many features so that the user does not have to talk over the phone or chat via the internet with customer support, or even to physically go to the bank branch. When a bank releases its own online banking app for smartphones, the customer is likely to download it …

Artificial Intelligence In Pursuit-Evasion Games, Specifically In The Scotland Yard Game, Arif M. Alamri

Theses and Dissertations

This research provides a heuristic algorithm for the detectives, who try to collectively capture a criminal known as Mr. X, in the Scotland Yard pursuer-evasion game. In Scotland Yard, a team of detectives attempts to converge on and capture a criminal known as Mr. X. The heuristic algorithm developed in this thesis is designed to emulate human strategies when playing the game. The algorithm uses the current state of the board at each time step, including the current positions of the detectives as well as the last known position of Mr. X. The heuristic algorithm then analyses all of the …

A Methodology To Identify Alternative Suitable Nosql Data Models Via Observation Of Relational Database Interactions, Paul M. Beach

Theses and Dissertations

The effectiveness and performance of data-intensive applications are influenced by the suitability of the data models upon which they are built. The relational data model has been the de facto data model underlying most database systems since the 1970’s. However, the recent emergence of NoSQL data models have provided users with alternative ways of storing and manipulating data. Previous research has demonstrated the potential value in applying NoSQL data models in non-distributed environments. However, knowing when to apply these data models has generally required inputs from system subject matter experts to make this determination. This research, sponsored by the Air …

Improving Closely Spaced Dim Object Detection Through Improved Multiframe Blind Deconvolution, Ronald M. Aung

Theses and Dissertations

This dissertation focuses on improving the ability to detect dim stellar objects that are in close proximity to a bright one, through statistical image processing using short exposure images. The goal is to improve the space domain awareness capabilities with the existing infrastructure. Two new algorithms are developed. The first one is through the Neighborhood System Blind Deconvolution where the data functions are separated into the bright object, the neighborhood system, and the background functions. The second one is through the Dimension Reduction Blind Deconvolution, where the object function is represented by the product of two matrices. Both are designed …

Physics-Constrained Hyperspectral Data Exploitation Across Diverse Atmospheric Scenarios, Nicholas M. Westing

Theses and Dissertations

Hyperspectral target detection promises new operational advantages, with increasing instrument spectral resolution and robust material discrimination. Resolving surface materials requires a fast and accurate accounting of atmospheric effects to increase detection accuracy while minimizing false alarms. This dissertation investigates deep learning methods constrained by the processes governing radiative transfer to efficiently perform atmospheric compensation on data collected by long-wave infrared (LWIR) hyperspectral sensors. These compensation methods depend on generative modeling techniques and permutation invariant neural network architectures to predict LWIR spectral radiometric quantities. The compensation algorithms developed in this work were examined from the perspective of target detection performance using …

Deep Learning Techniques For Structural Response Prediction During Strong Ground Motions, Ahmed A. Torky, Susumu Ohno Prof., Toshide Kashima Prof.

Civil Engineering

In this paper, deep learning techniques are applied to predict a building’s structural response to strong ground motions. Data from sensors near and inside structures measure accelerations during strong ground motions. The Building Research Institute (BRI) ANX building, an eight-story structure, has experienced major earthquakes since 1998. Sensors in the building provide and accumulate big data of historic events. Changes of the natural frequency of the ANX building from the big data is initially quantified. The time-series data of the historic events can be used to predict future response to future events using deep learning models rapidly. Although previous literature …

Machine Learning Applications For Drug Repurposing, Hansaim Lim

Dissertations, Theses, and Capstone Projects

The cost of bringing a drug to market is astounding and the failure rate is intimidating. Drug discovery has been of limited success under the conventional reductionist model of one-drug-one-gene-one-disease paradigm, where a single disease-associated gene is identified and a molecular binder to the specific target is subsequently designed. Under the simplistic paradigm of drug discovery, a drug molecule is assumed to interact only with the intended on-target. However, small molecular drugs often interact with multiple targets, and those off-target interactions are not considered under the conventional paradigm. As a result, drug-induced side effects and adverse reactions are often neglected …

Unclonable Secret Keys, Marios Georgiou

Dissertations, Theses, and Capstone Projects

We propose a novel concept of securing cryptographic keys which we call “Unclonable Secret Keys,” where any cryptographic object is modified so that its secret key is an unclonable quantum bit-string whereas all other parameters such as messages, public keys, ciphertexts, signatures, etc., remain classical. We study this model in the authentication and encryption setting giving a plethora of definitions and positive results as well as several applications that are impossible in a purely classical setting.

In the authentication setting, we define the notion of one-shot signatures, a fundamental element in building unclonable keys, where the signing key not only …

Role Of Influence In Complex Networks, Nur Dean

Dissertations, Theses, and Capstone Projects

Game theory is a wide ranging research area; that has attracted researchers from various fields. Scientists have been using game theory to understand the evolution of cooperation in complex networks. However, there is limited research that considers the structure and connectivity patterns in networks, which create heterogeneity among nodes. For example, due to the complex ways most networks are formed, it is common to have some highly “social” nodes, while others are highly isolated. This heterogeneity is measured through metrics referred to as “centrality” of nodes. Thus, the more “social” nodes tend to also have higher centrality.

In this thesis, …

Large Signal Modeling For Llc Resonant Converter, Zheng Kai, Jianbing Li, Zhou Dongfang, Li Kai, Songzhen Zhang

Journal of System Simulation

Abstract: The large-signal model modeling issue of LLC resonant converter was dealt with by using averaged large-signal modeling method, on the basis of a full-bridge LLC voltage-multiplying resonant converter. The SSOC (self-sustained oscillation controller) was analyzed, and the operation mode of LLC resonant converter under the SSOC mode was investigated. The nonlinear state space model of LLC resonant converter was established by analyzing linear approximation of nonlinear terms and harmonic balance, and the averaged large-signal model of LLC resonant converter was proposed. Two curves of large signal model and nonlinear state space model were proved to be identical …

Construction Of Virtual Atmospheric Environment Based On Mm5 And Sedris, Lianlei Lin, Ding Wei

Journal of System Simulation

Abstract: The new virtual test for virtual atmosphere environment puts forward higher requirements, such as providing complex atmospheric environment which includes a variety of atmospheric parameters and improving usability and reusability of atmospheric environment data. A method to construct virtual atmospheric environment based on MM5 and SEDRIS was proposed which met the above demands. The original data of complex atmospheric environment in any region and scale was generated by using MM5 model. According to the characteristics as well as the dynamic relationship between time and space of grid of atmospheric data, SEDRIS was selected to regulate appropriate DRM class, SRF …