Computer Engineering Commons^™

A Probabilistic Machine Learning Framework For Cloud Resource Selection On The Cloud, Syeduzzaman Khan

University of the Pacific Theses and Dissertations

The execution of the scientific applications on the Cloud comes with great flexibility, scalability, cost-effectiveness, and substantial computing power. Market-leading Cloud service providers such as Amazon Web service (AWS), Azure, Google Cloud Platform (GCP) offer various general purposes, memory-intensive, and compute-intensive Cloud instances for the execution of scientific applications. The scientific community, especially small research institutions and undergraduate universities, face many hurdles while conducting high-performance computing research in the absence of large dedicated clusters. The Cloud provides a lucrative alternative to dedicated clusters, however a wide range of Cloud computing choices makes the instance selection for the end-users. This thesis …

Brexit Election: Forecasting A Conservative Party Victory Through The Pound Using Arima And Facebook's Prophet, James Usher, Pierpaolo Dondio

Conference papers

On the 30th October, 2019, the markets watched as British Prime Minister, Boris Johnson, took a massive political gamble to call a general election to break the Withdrawal Agreement stalemate in the House of Commons to “Get BREXIT Done”. The pound had been politically sensitive owing to BREXIT uncertainty. With the polls indicating a Conservative win on 4thDecember, 2019, the margin of victory could be observed through increases in the pound. The outcome of a Conservative party victory would benefit the pound by removing the current market turbulence. We look to provide a short-term forecast of the pound. Our approach …

Brain Disease Detection From Eegs: Comparing Spiking And Recurrent Neural Networks For Non-Stationary Time Series Classification, Hristo Stoev

Dissertations

Modeling non-stationary time series data is a difficult problem area in AI, due to the fact that the statistical properties of the data change as the time series progresses. This complicates the classification of non-stationary time series, which is a method used in the detection of brain diseases from EEGs. Various techniques have been developed in the field of deep learning for tackling this problem, with recurrent neural networks (RNN) approaches utilising Long short-term memory (LSTM) architectures achieving a high degree of success. This study implements a new, spiking neural network-based approach to time series classification for the purpose of …

Content-Based Filtering Recommendation Approach To Label Irish Legal Judgements, Sandesh Gangadhar

Dissertations

Machine learning approaches are applied across several domains to either simplify or automate tasks which directly result in saved time or cost. Text document labelling is one such task that requires immense human knowledge about the domain and efforts to review, understand and label the documents. The company Stare Decisis summarises legal judgements and labels them as they are made available on Irish public legal source www.courts.ie. This research presents a recommendation-based approach to reduce the time for solicitors at Stare Decisis by reducing many numbers of available labels to pick from to a concentrated few that potentially contains the …

A Discrimination Aware Model To Predict Childhood Literacy Levels, Kate Byrne

Dissertations

It is illegal in Ireland to discriminate in the provision of education on the basis of multiple characteristics including gender, race and religion. While the increased use of machine learning models can open multiple avenues to identify early intervention strategies in education, caution must be exercised to ensure that any intervention does not discriminate with respect to a protected class. Poor literacy in childhood can have long term effects as the child ages, including on employment and mental health outcomes. Early intervention is key in mitigating this. In this dissertation, a model was created that predicted the outcome of a …

Optimization Of Home Mortgage Mover Predictive Model Applying Geo-Spatial Analysis And Machine Learning Techniques, Natalia Riscovaia

Dissertations

In the last decade digital innovations and online banking services have significantly changed customers banking preferences and behaviour. Banking industry is going through the changes and developments in the provision of banking services that are affecting the structure and the organization of the bank network. However, private home loan, referred as Home Mortgage hereinafter, continue to remain among the products, that customers prefer to have personal interaction about with professional advisors prior making the decision to apply for the loan with financial institution.

Synthetic Data Generation Using Wasserstein Conditional Gans With Gradient Penalty (Wcgans-Gp), Manhar Singh Walia

Dissertations

With data protection requirements becoming stricter, the data privacy has become increasingly important and more crucial than ever. This has led to restrictions on the availability and dissemination of real-world datasets. Synthetic data offers a viable solution to overcome barriers of data access and sharing. Existing data generation methods require a great deal of user-defined rules, manual interactions and domainspecific knowledge. Moreover, they are not able to balance the trade-off between datausability and privacy. Deep learning based methods like GANs have seen remarkable success in synthesizing images by automatically learning the complicated distributions and patterns of real data. But they …

Improving Transfer Learning For Use In Multi-Spectral Data, Yuvraj Sharma

Dissertations

Recently Nasa as well as the European Space Agency have made observational satellites images public. The main reason behind opening it to public is to foster research among university students and corporations alike. Sentinel is a program by the European Space Agency which has plans to release a series of seven satellites in lower earth orbit for observing land and sea patterns. Recently huge datasets have been made public by the Sentinel program. Many advancements have been made in the field of computer vision in the last decade. Krizhevsky, Sutskever & Hinton, 2012, revolutionized the field of image analysis by …

Exploration Of Approaches To Arabic Named Entity Recognition, Husamelddin Balla, Sarah Jane Delany

Conference papers

Abstract. The Named Entity Recognition (NER) task has attracted significant attention in Natural Language Processing (NLP) as it can enhance the performance of many NLP applications. In this paper, we compare English NER with Arabic NER in an experimental way to investigate the impact of using different classifiers and sets of features including language-independent and language-specific features. We explore the features and classifiers on five different datasets. We compare deep neural network architectures for NER with more traditional machine learning approaches to NER. We discover that most of the techniques and features used for English NER perform well on Arabic …

Android Compcache Based On Graphics Processing Unit, Muder Almi'ani, Abdu Razaque, Saleh Atiewi, Mohammed Alweshah, Ayman Al-Dmour, Basel Magableh

Conference papers

Android systems have been successfully developed to meet the demands of users. The following four methods are used in Android systems for memory management: backing swap, CompCache, traditional Linux swap, and low memory killer. These memory management methods are fully functioning.
However, Android phones cannot swap memory into solid-state drives, thus slowing the processor and reducing storage lifetime. In addition, the compression and decompression processes consume additional energy and latency. Therefore, the CompCache requires an extension. An extended Android CompCache using a graphics processing unit to compress and decompress memory pages on demand and reduce the latency is introduced in …

Deepmag+ : Sniffing Mobile Apps In Magnetic Field Through Deep Learning, Rui Ning, Cong Wang, Chunsheng Xin, Jiang Li, Hongyi Wu

Electrical & Computer Engineering Faculty Publications

This paper reports a new side-channel attack to smartphones using the unrestricted magnetic sensor data. We demonstrate that attackers can effectively infer the Apps being used on a smartphone with an accuracy of over 80%, through training a deep Convolutional Neural Networks (CNN). Various signal processing strategies have been studied for feature extractions, including a tempogram based scheme. Moreover, by further exploiting the unrestricted motion sensor to cluster magnetometer data, the sniffing accuracy can increase to as high as 98%. To mitigate such attacks, we propose a noise injection scheme that can effectively reduce the App sniffing accuracy to only …

A Saliency-Driven Video Magnifier For People With Low Vision, Ali Selman Aydin, Shirin Feiz, Iv Ramakrishnan, Vikas Ashok

Computer Science Faculty Publications

Consuming video content poses significant challenges for many screen magnifier users, which is the “go to” assistive technology for people with low vision. While screen magnifier software could be used to achieve a zoom factor that would make the content of the video visible to low-vision users, it is oftentimes a major challenge for these users to navigate through videos. Towards making videos more accessible for low-vision users, we have developed the SViM video magnifier system [6]. Specifically, SViM consists of three different magnifier interfaces with easy-to-use means of interactions. All three interfaces are driven by visual saliency as a …

Applying Artificial Intelligence To Medical Data, Shaikh Shiam Rahman

College of Graduate Studies: Theses & Dissertations

Machine learning, data mining, and deep learning has become the methodology of choice for analyzing medical data and images. In this study, we implemented three different machine learning techniques to medical data and image analysis. Our first study was to implement different log base entropy for a decision tree algorithm. Our results suggested that using a higher log base for the dataset with mostly categorical attributes with three or more categories for each attribute can obtain a higher accuracy. For the second study, we analyzed mental health data tuning the parameters of the decision tree (splitting method, depth and entropy). …

A Sample Weight And Adaboost Cnn-Based Coarse To Fine Classification Of Fruit And Vegetables At A Supermarket Self-Checkout, Khurram Hameed, Douglas Chai, Alexander Rassau

Research outputs 2014 to 2021

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. The physical features of fruit and vegetables make the task of vision-based classification of fruit and vegetables challenging. The classification of fruit and vegetables at a supermarket self-checkout poses even more challenges due to variable lighting conditions and human factors arising from customer interactions with the system along with the challenges associated with the colour, texture, shape, and size of a fruit or vegetable. Considering this complex application, we have proposed a progressive coarse to fine classification technique to classify fruit and vegetables at supermarket checkouts. The image and weight of …

Proactive Content Caching In Future Generation Communication Networks: Energy And Security Considerations, Muhammad Ishtiaque Aziz Zahed

Theses: Doctorates and Masters

The proliferation of hand-held devices and Internet of Things (IoT) applications has heightened demand for popular content download. A high volume of content streaming/downloading services during peak hours can cause network congestion. Proactive content caching has emerged as a prospective solution to tackle this congestion problem. In proactive content caching, data storage units are used to store popular content in helper nodes at the network edge. This contributes to a reduction of peak traffic load and network congestion.

However, data storage units require additional energy, which offers a challenge to researchers that intend to reduce energy consumption up to 90% …

Application Of Advanced Algorithms And Statistical Techniques For Weed-Plant Discrimination, Saman Akbar Zadeh

Theses: Doctorates and Masters

Precision agriculture requires automated systems for weed detection as weeds compete with the crop for water, nutrients, and light. The purpose of this study is to investigate the use of machine learning methods to classify weeds/crops in agriculture. Statistical methods, support vector machines, convolutional neural networks (CNNs) are introduced, investigated and optimized as classifiers to provide high accuracy at high vehicular speed for weed detection.

Initially, Support Vector Machine (SVM) algorithms are developed for weed-crop discrimination and their accuracies are compared with a conventional data-aggregation method based on the evaluation of discrete Normalised Difference Vegetation Indices (NDVIs) at two different …

Impact Of Direction Parameter In Performance Of Modified Aodv In Vanet, Afsana Ahamed, Hamid Vakilzadian

Department of Electrical and Computer Engineering: Faculty Publications

A vehicular ad hoc network (VANET) is a technology in which moving cars are used as routers (nodes) to establish a reliable mobile communication network among the vehicles. Some of the drawbacks of the routing protocol, Ad hoc On-Demand Distance Vector (AODV), associated with VANETs are the end-to-end delay and packet loss. We modified the AODV routing protocols to reduce the number of route request (RREQ) and route reply (RREP) messages by adding direction parameters and two-step filtering. The two-step filtering process reduces the number of RREQ and RREP packets, reduces the packet overhead, and helps to select the stable …

Far-Field Transient Absorption Nanoscopy With Sub-50 Nm Optical Super-Resolution, Yali Bi, Chi Yang, Lei Tong, Haozheng Li, Boyu Yu, Shuai Yan, Guang Yang, Meng Deng, Yi Wang, Wei Bao, Lei Ye, Ping Wang

Department of Electrical and Computer Engineering: Faculty Publications

Nanoscopic imaging or characterizing is the mainstay of the development of advanced materials. Despite great progress in electronic and atomic force microscopies, label-free and far-field characterization of materials with deep sub- wavelength spatial resolution has long been highly desired. Herein, we demonstrate far-field super-resolution transient absorption (TA) imaging of two-dimensional material with a spatial resolution of sub-50 nm. By introducing a donut- shaped blue saturation laser, we effectively suppress the TA transition driven by near-infrared (NIR) pump–probe photons, and push the NIR-TA microscopy to sub-diffraction-limited resolution. Specifically, we demonstrate that our method can image the individual nano-grains in graphene with …

Precursor-Surface Interactions Revealed During Plasma-Enhanced Atomic Layer Deposition Of Metal Oxide Thin Films By In-Situ Spectroscopic Ellipsometry, Ufuk Kilic, Alyssa Mock, Derek Sekora, Simeon Gilbert, Shah R. Valloppilly, Natale J. Ianno, Marjorie Langell, Eva Schubert, Mathias Schubert

Department of Electrical and Computer Engineering: Faculty Publications

We find that a five-phase (substrate, mixed native oxide and roughness interface layer, metal oxide thin film layer, surface ligand layer, ambient) model with two-dynamic (metal oxide thin film layer thickness and surface ligand layer void fraction) parameters (dynamic dual box model) is sufficient to explain in-situ spectroscopic ellipsometry data measured within and across multiple cycles during plasma-enhanced atomic layer deposition of metal oxide thin films. We demonstrate our dynamic dual box model for analysis of in-situ spectroscopic ellipsometry data in the photon energy range of 0.7–3.4 eV measured with time resolution of few seconds over large numbers of cycles …

Transcriptional Analysis Of Cleft Palate In Tgfβ3 Mutant Mice, J. Liu, S.K. Chanumolu, K.M. White, M. Albahrani, Hasan H. Otu, A. Nawshad

Department of Electrical and Computer Engineering: Faculty Publications

Cleft palate (CP) is one of the most common craniofacial birth defects, impacting about 1 in 800 births in the USA. Tgf-β3 plays a critical role in regulating murine palate development, and Tgf-β3 null mutants develop cleft palate with 100% penetrance. In this study, we compared global palatal transcriptomes of wild type (WT) and Tgf-β3 −/− homozygous (HM) mouse embryos at the crucial palatogenesis stages of E14.5, and E16.5, using RNA-seq data. We found 1,809 and 2,127 differentially expressed genes at E16.5 vs. E14.5 in the WT and HM groups, respectively (adjusted p < 0.05; |fold change|> 2.0). We focused on the genes that …