Digital Commons Network^™

Prediction And Optimal Scheduling Of Advertisements In Linear Television, Mark J. Panaggio, Pak-Wing Fok, Ghan S. Bhatt, Simon Burhoe, Michael Capps, Christina J. Edholm, Fadoua El Moustaid, Tegan Emerson, Star-Lena Estock, Nathan Gold, Ryan Halabi, Madelyn Houser, Peter R. Kramer, Hsuan-Wei Lee, Qingxia Li, Weiqiang Li, Dan Lu, Yuzhou Qian, Louis F. Rossi, Deborah Shutt, Vicky Chuqiao Yang, Yingxiang Zhou

Mathematical Sciences Faculty Research

Advertising is a crucial component of marketing and an important way for companies to raise awareness of goods and services in the marketplace. Advertising campaigns are designed to convey a marketing image or message to an audience of potential consumers and television commercials can be an effective way of transmitting these messages to a large audience. In order to meet the requirements for a typical advertising order, television content providers must provide advertisers with a predetermined number of "impressions" in the target demographic. However, because the number of impressions for a given program is not known a priori and because …

Methodology To Predict Daily Groundwater Levels By The Implementation Of Machine Learning And Crop Models, Sandra Milena Guzman Gutierrez

Theses and Dissertations

The continuous decline of groundwater levels caused by variations in climatic conditions and crop water demands is an increased concern for the agricultural community. It is necessary to understand the factors that control these changes in groundwater levels so that we can better address declines and develop improved conservation practices that will lead to a more sustainable use of water. In this study, two machine learning techniques namely support vector regression (SVR) and the nonlinear autoregressive with exogenous inputs (NARX) neural network were implemented to predict daily groundwater levels in a well located in the Mississippi Delta Region (MDR). Results …

Learning From Data: Plant Breeding Applications Of Machine Learning, Alencar Xavier

Open Access Dissertations

Increasingly, new sources of data are being incorporated into plant breeding pipelines. Enormous amounts of data from field phenomics and genotyping technologies places data mining and analysis into a completely different level that is challenging from practical and theoretical standpoints. Intelligent decision-making relies on our capability of extracting from data useful information that may help us to achieve our goals more efficiently. Many plant breeders, agronomists and geneticists perform analyses without knowing relevant underlying assumptions, strengths or pitfalls of the employed methods. The study endeavors to assess statistical learning properties and plant breeding applications of supervised and unsupervised machine learning …

Teacher + Technology = Blended Learning: How Important Is The Teacher In This Equation?, Catherine A. Doom

Department of Educational Administration: Dissertations, Theses, and Student Research

This mixed method study investigates teacher belief, knowledge action and technology adoption rate in a blended learning setting and the impact those variables have on student comprehension measures. Surveys and data collected from the blended learning program were used to gather data in pursuit of answers to the following research questions: (a) Does the teacher’s adoption of technology rate impact reading comprehension growth in a blended learning environment? (b) Does the teacher’s knowledge, belief and action in a blended learning environment impact student reading comprehension growth? (c) Does the teacher’s adoption of technology rate combined with the teacher’s knowledge, attitude, …

Learning In The Real World: Constraints On Cost, Space, And Privacy, Matt J. Kusner

McKelvey School of Engineering Graduate Student Theses & Dissertations

The sheer demand for machine learning in fields as varied as: healthcare, web-search ranking, factory automation, collision prediction, spam filtering, and many others, frequently outpaces the intended use-case of machine learning models. In fact, a growing number of companies hire machine learning researchers to rectify this very problem: to tailor and/or design new state-of-the-art models to the setting at hand.

However, we can generalize a large set of the machine learning problems encountered in practical settings into three categories: cost, space, and privacy. The first category (cost) considers problems that need to balance the accuracy of a machine learning model …

Lidar-Assisted Extraction Of Old Growth Baldcypress Stands Along The Black River Of North Carolina, Weston Pierce Murch

Graduate Theses and Dissertations

The remnants of ancient baldcypress forests continue to grow across the Southeastern United States. These long lived trees are invaluable for biodiversity along riverine ecosystems, provide habitat to a myriad of animal species, and augment the proxy climate record for North America. While extensive logging of the areas along the Black River in North Carolina has mostly decimated ancient forests of many species including the baldcypress, conservation efforts from The Nature Conservancy and other partners are under way. In order to more efficiently find and study these enduring stands of baldcypress, some of which are estimated to be more than …

Optcon: An Adaptable Sla-Aware Consistency Tuning Framework For Quorum-Based Stores, Subhajit Sidhanta, Wojciech Golab, Supratik Mukhopadhyay, Saikat Basu

Faculty Publications

Users of distributed datastores that employquorum-based replication are burdened with the choice of asuitable client-centric consistency setting for each storage operation. The above matching choice is difficult to reason about asit requires deliberating about the tradeoff between the latencyand staleness, i.e., how stale (old) the result is. The latencyand staleness for a given operation depend on the client-centricconsistency setting applied, as well as dynamic parameters such asthe current workload and network condition. We present OptCon, a novel machine learning-based predictive framework, that canautomate the choice of client-centric consistency setting underuser-specified latency and staleness thresholds given in the servicelevel agreement (SLA). …

Application Of Machine Learning To Prediction Of Vegetation Health, Emily Burchfield, John J. Nay, Jonathan Gilligan

Environment and Society Faculty Publications

This project applies machine learning techniques to remotely sensed imagery to train and validate predictive models of vegetation health in Bangladesh and Sri Lanka. For both locations, we downloaded and processed eleven years of imagery from multiple MODIS datasets which were combined and transformed into two-dimensional matrices. We applied a gradient boosted machines model to the lagged dataset values to forecast future values of the Enhanced Vegetation Index (EVI). The predictive power of raw spectral data MODIS products were compared across time periods and land use categories. Our models have significantly more predictive power on held-out datasets than a baseline. …

Location Inference Of Social Media Posts At Hyper-Local Scale, Brian D. Mcclanahan

Master's Theses

This paper describes an approach to infer the location of a social media post at a hyper-local scale based on its content, conditional to the knowledge that the post originates from a larger area such as a city or even a state. The approach comprises three components: (i) a discriminative classifier, namely, Logistic Regression (LR) which selects from a set of most probable sub-regions from where 1 a post might have originated; (ii) a clustering technique, namely, k-means, that adaptively partitions the larger geographic region into sub-regions based on the density of the posts; and (iii) a range of techniques …

Spectral Target Detecting Using Schroedinger Eigenmaps, Leidy P. Dorado-Munoz

Theses

Applications of optical remote sensing processes include environmental monitoring, military monitoring, meteorology, mapping, surveillance, etc. Many of these tasks include the detection of specific objects or materials, usually few or small, which are surrounded by other materials that clutter the scene and hide the relevant information. This target detection process has been boosted lately by the use of hyperspectral imagery (HSI) since its high spectral dimension provides more detailed spectral information that is desirable in data exploitation. Typical spectral target detectors rely on statistical or geometric models to characterize the spectral variability of the data. However, in many cases these …

“When ‘Bad’ Is ‘Good’”: Identifying Personal Communication And Sentiment In Drug-Related Tweets, Raminta Daniulaityte, Lu Chen, Francois R. Lamy, Robert G. Carlson, Krishnaprasad Thirunarayan, Amit Sheth

Population and Public Health Sciences Faculty Publications

Background: To harness the full potential of social media for epidemiological surveillance of drug abuse trends, the field needs a greater level of automation in processing and analyzing social media content. Objectives: The objective of the study is to describe the development of supervised machine-learning techniques for the eDrugTrends platform to automatically classify tweets by type/source of communication (personal, official/media, retail) and sentiment (positive, negative, neutral) expressed in cannabis- and synthetic cannabinoid-related tweets. Methods: Tweets were collected using Twitter streaming Application Programming Interface and filtered through the eDrugTrends platform using keywords related to cannabis, marijuana edibles, marijuana concentrates, and synthetic …

Designing Human-Centered Collective Intelligence, Ivor Addo

Dissertations (1934 -)

Human-Centered Collective Intelligence (HCCI) is an emergent research area that seeks to bring together major research areas like machine learning, statistical modeling, information retrieval, market research, and software engineering to address challenges pertaining to deriving intelligent insights and solutions through the collaboration of several intelligent sensors, devices and data sources. An archetypal contextual CI scenario might be concerned with deriving affect-driven intelligence through multimodal emotion detection sources in a bid to determine the likability of one movie trailer over another. On the other hand, the key tenets to designing robust and evolutionary software and infrastructure architecture models to address cross-cutting …

Machine Learning Methods For Brain Image Analysis, Ahmed Fakhry

Computer Science Theses & Dissertations

Understanding how the brain functions and quantifying compound interactions between complex synaptic networks inside the brain remain some of the most challenging problems in neuroscience. Lack or abundance of data, shortage of manpower along with heterogeneity of data following from various species all served as an added complexity to the already perplexing problem. The ability to process vast amount of brain data need to be performed automatically, yet with an accuracy close to manual human-level performance. These automated methods essentially need to generalize well to be able to accommodate data from different species. Also, novel approaches and techniques are becoming …

Machine Learning Methods For Medical And Biological Image Computing, Rongjian Li

Computer Science Theses & Dissertations

Medical and biological imaging technologies provide valuable visualization information of structure and function for an organ from the level of individual molecules to the whole object. Brain is the most complex organ in body, and it increasingly attracts intense research attentions with the rapid development of medical and bio-logical imaging technologies. A massive amount of high-dimensional brain imaging data being generated makes the design of computational methods for efficient analysis on those images highly demanded. The current study of computational methods using hand-crafted features does not scale with the increasing number of brain images, hindering the pace of scientific discoveries …

Short-Term Building Energy Model Recommendation System: A Meta-Learning Approach, Can Cui, Teresa Wu, Mengqi Hu, Jeffery D. Weir, Xiwang Li

Faculty Publications

High-fidelity and computationally efficient energy forecasting models for building systems are needed to ensure optimal automatic operation, reduce energy consumption, and improve the building’s resilience capability to power disturbances. Various models have been developed to forecast building energy consumption. However, given buildings have different characteristics and operating conditions, model performance varies. Existing research has mainly taken a trial-and-error approach by developing multiple models and identifying the best performer for a specific building, or presumed one universal model form which is applied on different building cases. To the best of our knowledge, there does not exist a generalized system framework which …

Determining The Effectiveness Of Soil Treatment On Plant Stress Using Smart-Phone Cameras, Anurag Panwar

USF Tampa Graduate Theses and Dissertations

Plants are vital to the health of our biosphere, and effectively sustaining their growth is fundamental to the existence of life on this planet. A critical aspect, which decides the sustainability of plant growth is the quality of soil. All other things being fixed, the quality of soil greatly impacts the plant stress, which in turn impacts overall health. Although plant stress manifests in many ways, one of the clearest indicators are colors of the leaves. In this thesis, we conducted an experimental study in a greenhouse for detecting plant stress caused by nutrient deficienceies in soil using smartphone cameras, …

Answer-Type Prediction For Visual Question Answering, Kushal Kafle, Christopher Kanan

Presentations and other scholarship

Recently, algorithms for object recognition and related tasks have become sufficiently proficient that new vision tasks can now be pursued. In this paper, we build a system capable of answering open-ended text-based questions about images, which is known as Visual Question Answering (VQA). Our approach’s key insight is that we can predict the form of the answer from the question. We formulate our solution in a Bayesian framework. When our approach is combined with a discriminative model, the combined model achieves state-of-the-art results on four benchmark datasets for open-ended VQA: DAQUAR, COCO-QA, The VQA Dataset, and Visual7W.

Machine Learning For Disease Prediction, Abraham Jacob Frandsen

Theses and Dissertations

Millions of people in the United States alone suffer from undiagnosed or late-diagnosed chronic diseases such as Chronic Kidney Disease and Type II Diabetes. Catching these diseases earlier facilitates preventive healthcare interventions, which in turn can lead to tremendous cost savings and improved health outcomes. We develop algorithms for predicting disease occurrence by drawing from ideas and techniques in the field of machine learning. We explore standard classification methods such as logistic regression and random forest, as well as more sophisticated sequence models, including recurrent neural networks. We focus especially on the use of medical code data for disease prediction, …

Exploring Data Mining Techniques For Tree Species Classification Using Co-Registered Lidar And Hyperspectral Data, Julia K. Marrs

Theses and Dissertations

NASA Goddard’s LiDAR, Hyperspectral, and Thermal imager provides co-registered remote sensing data on experimental forests. Data mining methods were used to achieve a final tree species classification accuracy of 68% using a combined LiDAR and hyperspectral dataset, and show promise for addressing deforestation and carbon sequestration on a species-specific level.

Mining Knowledge Bases For Question & Answers Websites, Eduardo Coelho De Lima

Theses

We studied the problem of searching answers for questions on a Question-and-Answer Website from knowledge bases. A number of research efforts had been developed using Stack Overflow data, which is available for the public. Surprisingly, only a few papers tried to improve the search for better answers. Furthermore, current approaches for searching a Question-and-Answer Website are usually limited to the question database, which is usually the website own content. We showed it is feasible to use knowledge bases as sources for answers. We implemented both vector-space and topic-space representations for our datasets and compared these distinct techniques. Finally, we proposed …

Revelation Of Yin-Yang Balance In Microbial Cell Factories By Data Mining, Flux Modeling, And Metabolic Engineering, Gang Wu

McKelvey School of Engineering Graduate Student Theses & Dissertations

The long-held assumption of never-ending rapid growth in biotechnology and especially in synthetic biology has been recently questioned, due to lack of substantial return of investment. One of the main reasons for failures in synthetic biology and metabolic engineering is the metabolic burdens that result in resource losses. Metabolic burden is defined as the portion of a host cells resources either energy molecules (e.g., NADH, NADPH and ATP) or carbon building blocks (e.g., amino acids) that is used to maintain the engineered components (e.g., pathways). As a result, the effectiveness of synthetic biology tools heavily dependents on cell capability to …

Surface Chemistry With Machine Learning And Quantum Mechanics, Venkatesh Botu

Doctoral Dissertations

Surface chemistry is a phenomenon manifesting itself in several key areas; catalysis, materials fabrication, and emissions mitigation, to name a few. At the present time, atomistic computational driven efforts to study such processes are dominated by models based on quantum mechanics. Their flexibility in studying diverse chemistries, along with the ability to predict accurate thermodynamic and kinetic insights of surface processes, makes them increasingly popular. From ultra-low temperature and pressure to normal operating conditions these methods are now commonly utilized. Nevertheless, the computational burden inherent in the method renders it insufficient to keep up with the current need for quick …

Evolutionary Weights For Random Subspace Learning, Andre Lobato Ramos

Theses

Ensemble learning is a widely used technique in Data Mining, this method allows us to aggregate models to reduce prediction error. There are many methods on how to perform model aggregation, one of them is known as Random Subspace Learning, which consists of building subspace of the feature space where we want to create our models. The task of selecting good subspaces and in turn produce good models for better prediction can be a daunting one, so we want to propose a new method to accomplish such a task. This proposed method allows for an automated data-driven way to attribute …

A Comparative Approach To Question Answering Systems, Josue Balandrano Coronel

Theses and Dissertations

In this paper I will analyze three different algorithms and approaches to implement Question Answering Systems (QA-Systems). I will analyze the efficiency, strengths, and weaknesses of multiple algorithms by explaining them in detail and comparing them with each other. The overarching aim of this thesis is to explore ideas that can be used to create a truly open context QA-System. Open context QA-Systems remain an open problem.

The various algorithms and approaches presented in this work will be focused on complex questions. Complex questions are usually verbose and the context of the question is equally important to answer the query …

Assessing Gender Inequality From Large Scale Online Student Reviews, Souradeep Sinha

Dissertations - ALL

Career growth in academia is often dependent on student reviews of university profes-

sors. A growing concern is how evaluation of teaching has been affected by gender biases

throughout the reviewing process. However, pinpointing the exact causes and consequen-

tial effects of this form of gender inequality has been a hard task.

Current work focusses on university-wide student reviewing system, that depends on

objective responses on a Likert scale to measure various aspects of an instructor’s qual-

ity. Through our work, we access online student review data which are not limited by

geographies, universities, or disciplines.

Thereafter, we come up …

Exploring Privacy Leakage From The Resource Usage Patterns Of Mobile Apps, Amin Rois Sinung Nugroho

Graduate Theses and Dissertations

Due to the popularity of smart phones and mobile apps, a potential privacy risk with the usage of mobile apps is that, from the usage information of mobile apps (e.g., how many hours a user plays mobile games in each day), private information about a user’s living habits and personal activities can be inferred. To assess this risk, this thesis answers the following research question: can the type of a mobile app (e.g., email, web browsing, mobile game, music streaming, etc.) used by a user be inferred from the resource (e.g., CPU, memory, network, etc.) usage patterns of the mobile …

Memory And Information Processing In Recurrent Neural Networks, Alireza Goudarzi, Sarah Marzen, Peter Banda, Guy Feldman, Matthew R. Lakin, Christof Teuscher, Darko Stefanovic

Electrical and Computer Engineering Faculty Publications and Presentations

Recurrent neural networks (RNN) are simple dynamical systems whose computational power has been attributed to their short-term memory. Short-term memory of RNNs has been previously studied analytically only for the case of orthogonal networks, and only under annealed approximation, and uncorrelated input. Here for the first time, we present an exact solution to the memory capacity and the task-solving performance as a function of the structure of a given network instance, enabling direct determination of the function-structure relation in RNNs. We calculate the memory capacity for arbitrary networks with exponentially correlated input and further related it to the performance of …

Gait And Tremor Monitoring System For Patients With Parkinson’S Disease Using Wearable Sensors, Shyam Vignesh Perumal

USF Tampa Graduate Theses and Dissertations

Typically, a Parkinson’s disease (PD) patient would display instances of tremor and bradykinesia (slowness of movement) at an early stage of the disease and later develop gait disturbances and postural instability. So, it is important to measure the tremor occurrences in subjects to detect the onset of PD. Also, it is equally essential to monitor the gait impairments that the patient displays, as the order at which the PD symptoms appear in subjects vary from one to another.

The primary goal of this thesis is to develop a monitoring system for PD patients using wearable sensors. To achieve that objective, …

Bridging Statistical Learning And Formal Reasoning For Cyber Attack Detection, Kexin Pei

Open Access Theses

Current cyber-infrastructures are facing increasingly stealthy attacks that implant malicious payloads under the cover of benign programs. Current attack detection approaches based on statistical learning methods may generate misleading decision boundaries when processing noisy data with such a mixture of benign and malicious behaviors. On the other hand, attack detection based on formal program analysis may lack completeness or adaptivity when modeling attack behaviors. In light of these limitations, we have developed LEAPS, an attack detection system based on supervised statistical learning to classify benign and malicious system events. Furthermore, we leverage control flow graphs inferred from the system event …

Learning Robust And Discriminative Manifold Representations For Pattern Recognition, Sriram Kumar

Theses

Face and object recognition find applications in domains such as biometrics, surveillance and human computer interaction. An important component in any recognition pipeline is to learn pertinent image representations that will help the system to discriminate one image class from another. These representations enable the system to learn a discriminative function that can classify a wide range of images. In practical situations, the images acquired are often corrupted with occlusions and noise. Thus, a robust and discriminative learning is necessary for good classification performance.

This thesis explores two scenarios where robust and discriminative manifold representations help recognize face and object …