Numerical Analysis and Scientific Computing Commons^™

Machine Learning Algorithms For Automated Satellite Snow And Sea Ice Detection, George Bonev

Dissertations, Theses, and Capstone Projects

The continuous mapping of snow and ice cover, particularly in the arctic and poles, are critical to understanding the earth and atmospheric science. Much of the world's sea ice and snow covers the most inhospitable places, making measurements from satellite-based remote sensors essential. Despite the wealth of data from these instruments many challenges remain. For instance, remote sensing instruments reside on-board different satellites and observe the earth at different portions of the electromagnetic spectrum with different spatial footprints. Integrating and fusing this information to make estimates of the surface is a subject of active research.

In response to these challenges, …

Machine Learning Based Protein Sequence To (Un)Structure Mapping And Interaction Prediction, Sumaiya Iqbal

LSU New Orleans Theses and Dissertations

Proteins are the fundamental macromolecules within a cell that carry out most of the biological functions. The computational study of protein structure and its functions, using machine learning and data analytics, is elemental in advancing the life-science research due to the fast-growing biological data and the extensive complexities involved in their analyses towards discovering meaningful insights. Mapping of protein’s primary sequence is not only limited to its structure, we extend that to its disordered component known as Intrinsically Disordered Proteins or Regions in proteins (IDPs/IDRs), and hence the involved dynamics, which help us explain complex interaction within a cell that …

Ancr—An Adaptive Network Coding Routing Scheme For Wsns With Different-Success-Rate Links †, Xiang Ji, Anwen Wang, Chunyu Li, Chun Ma, Yao Peng, Dajin Wang, Qingyi Hua, Feng Chen, Dingyi Fang

Department of Computer Science Faculty Scholarship and Creative Works

As the underlying infrastructure of the Internet of Things (IoT), wireless sensor networks (WSNs) have been widely used in many applications. Network coding is a technique in WSNs to combine multiple channels of data in one transmission, wherever possible, to save node’s energy as well as increase the network throughput. So far most works on network coding are based on two assumptions to determine coding opportunities: (1) All the links in the network have the same transmission success rate; (2) Each link is bidirectional, and has the same transmission success rate on both ways. However, these assumptions may not be …

Predictive Power And Validity Of Connectome Predictive Modeling: A Replication And Extension, Michael Wang, Joaquin Goni, Enrico Amico

The Summer Undergraduate Research Fellowship (SURF) Symposium

Neuroimaging, particularly functional magnetic resonance imaging (fMRI), is a rapidly growing research area and has applications ranging from disease classification to understanding neural development. With new advancements in imaging technology, researchers must employ new techniques to accommodate the influx of high resolution data sets. Here, we replicate a new technique: connectome-based predictive modeling (CPM), which constructs a linear predictive model of brain connectivity and behavior. CPM’s advantages over classic machine learning techniques include its relative ease of implementation and transparency compared to “black box” opaqueness and complexity. Is this method efficient, powerful, and reliable in the prediction of behavioral measures …

Improving The Accuracy For The Long-Term Hydrologic Impact Assessment (L-Thia) Model, Anqi Zhang, Lawrence Theller, Bernard A. Engel

The Summer Undergraduate Research Fellowship (SURF) Symposium

Urbanization increases runoff by changing land use types from less impervious to impervious covers. Improving the accuracy of a runoff assessment model, the Long-Term Hydrologic Impact Assessment (L-THIA) Model, can help us to better evaluate the potential uses of Low Impact Development (LID) practices aimed at reducing runoff, as well as to identify appropriate runoff and water quality mitigation methods. Several versions of the model have been built over time, and inconsistencies have been introduced between the models. To improve the accuracy and consistency of the model, the equations and parameters (primarily curve numbers in the case of this model) …

Optimization And Control Of Production Of Graphene, Atharva Hans, Nimish M. Awalgaonkar, Majed Alrefae, Ilias Bilionis, Timothy S. Fisher

The Summer Undergraduate Research Fellowship (SURF) Symposium

Graphene is a 2-dimensional element of high practical importance. Despite its exceptional properties, graphene’s real applications in industrial or commercial products have been limited. There are many methods to produce graphene, but none has been successful in commercializing its production. Roll-to-roll plasma chemical vapor deposition (CVD) is used to manufacture graphene at large scale. In this research, we present a Bayesian linear regression model to predict the roll-to-roll plasma system’s electrode voltage and current; given a particular set of inputs. The inputs of the plasma system are power, pressure and concentration of gases; hydrogen, methane, oxygen, nitrogen and argon. This …

Machine Learning In Xenon1t Analysis, Dillon A. Davis, Rafael F. Lang, Darryl P. Masson

The Summer Undergraduate Research Fellowship (SURF) Symposium

In process of analyzing large amounts of quantitative data, it can be quite time consuming and challenging to uncover populations of interest contained amongst the background data. Therefore, the ability to partially automate the process while gaining additional insight into the interdependencies of key parameters via machine learning seems quite appealing. As of now, the primary means of reviewing the data is by manually plotting data in different parameter spaces to recognize key features, which is slow and error prone. In this experiment, many well-known machine learning algorithms were applied to a dataset to attempt to semi-automatically identify known populations, …

Predicting Locations Of Pollution Sources Using Convolutional Neural Networks, Yiheng Chi, Nickolas D. Winovich, Guang Lin

The Summer Undergraduate Research Fellowship (SURF) Symposium

Pollution is a severe problem today, and the main challenge in water and air pollution controls and eliminations is detecting and locating pollution sources. This research project aims to predict the locations of pollution sources given diffusion information of pollution in the form of array or image data. These predictions are done using machine learning. The relations between time, location, and pollution concentration are first formulated as pollution diffusion equations, which are partial differential equations (PDEs), and then deep convolutional neural networks are built and trained to solve these PDEs. The convolutional neural networks consist of convolutional layers, reLU layers …

Structure-Force Field Generator For Molecular Dynamics Simulations, Carlos M. Patiño, Lorena Alzate, Alejandro Strachan

The Summer Undergraduate Research Fellowship (SURF) Symposium

Atomistic and molecular simulations have become an important research field due to the progress made in computer performance and the necessity of new and improved materials. Despite this, first principle simulations of large molecules are still not possible because the high computational time and resources required. Other methods, such as molecular dynamics, allow the simplification of calculations by defining energy terms to describe multiple atom interactions without compromising accuracy significantly. A group of these energy terms is called a force field, and each force field has its own descriptions and parameters. The objective of this project was to develop a …

Evaluating Relevance And Reliability Of Twitter Data For Risk Communication, Xiaohui Liu

Dissertations

While Twitter has been touted to provide up-to-date information about hazard events, the relevance and reliability of tweets is yet to be tested. This research examined the relevance and reliability of risk information extracted from Twitter during the 2013 Colorado floods using five different approaches. The first approach examined the relationship between tweet volume and precipitation amount. The second approach explored the relationship between geo-tagged tweets and degree of damage. In the third approach, the spatiotemporal distribution of tweets was compared with flood extent. In the fourth approach, risk information from tweets were compared with survey responses obtained in a …

Modeling Trajectories With Recurrent Neural Networks, Hao Wu, Ziyang Chen, Weiwei Sun, Baihua Zheng, Wei Wang

Research Collection School Of Computing and Information Systems

Modeling trajectory data is a building block for many smart-mobility initiatives. Existing approaches apply shallow models such as Markov chain and inverse reinforcement learning to model trajectories, which cannot capture the long-term dependencies. On the other hand, deep models such as Recurrent Neura lNetwork (RNN) have demonstrated their strength of modeling variable length sequences. However, directly adopting RNN to model trajectories is not appropriate because of the unique topological constraints faced by trajectories. Motivated by these findings, we design two RNN-based models which can make full advantage of the strength of RNN to capture variable length sequence and meanwhile to …

Deepfacade: A Deep Learning Approach To Facade Parsing, Hantang Liu, Jialiang Zhang, Jianke Zhu, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

The parsing of building facades is a key component to the problem of 3D street scenes reconstruction, which is long desired in computer vision. In this paper, we propose a deep learning based method for segmenting a facade into semantic categories. Man-made structures often present the characteristic of symmetry. Based on this observation, we propose a symmetric regularizer for training the neural network. Our proposed method can make use of both the power of deep neural networks and the structure of man-made architectures. We also propose a method to refine the segmentation results using bounding boxes generated by the Region …

Basket-Sensitive Personalized Item Recommendation, Duc Trong Le, Hady W. Lauw, Yuan Fang

Research Collection School Of Computing and Information Systems

Personalized item recommendation is useful in narrowing down the list of options provided to a user. In this paper, we address the problem scenario where the user is currently holding a basket of items, and the task is to recommend an item to be added to the basket. Here, we assume that items currently in a basket share some association based on an underlying latent need, e.g., ingredients to prepare some dish, spare parts of some device. Thus, it is important that a recommended item is relevant not only to the user, but also to the existing items in the …

Generating Cultural Personas From Social Data: A Perspective Of Middle Eastern Users, Salminen Joni, Sercan Sengün, Haewoon Kwak, Bernard Jansen, Jisun An, Soon-Gyo Jung, Sarah Vieweg, D. Fox Harrell

Research Collection School Of Computing and Information Systems

We conduct a mixed-method study to better understand the content consumption patterns of Middle Eastern social media users and to explore new ways to present online data by using automatic persona generation. First, we analyze millions of content interactions on YouTube to dynamically generate personas describing behavioral patterns of different demographic groups. Second, we analyze interview data on social media users in the Middle Eastern region to generate additional insights into the dynamically generated personas. Our findings provide insights into social media users in the Middle East, as well as present a novel methodology of using computational analysis and qualitative …

Multiplex Media Attention And Disregard Network Among 129 Countries, Haewoon Kwak, Jisun An

Research Collection School Of Computing and Information Systems

We built a multiplex media attention and disregard network (MADN) among 129 countries over 212 days. By characterizing the MADN from multiple levels, we found that it is formed primarily by skewed, hierarchical, and asymmetric relationships. Also, we found strong evidence that our news world is becoming a "global village." However, at the same time, unique attention blocks of the Middle East and North Africa (MENA) region, as well as Russia and its neighbors, still exist.

Personas For Content Creators Via Decomposed Aggregate Audience Statistics, Jisun An, Haewoon Kwak, Bernard J. Jansen

Research Collection School Of Computing and Information Systems

We propose a novel method for generating personas based on online user data for the increasingly common situation of content creators distributing products via online platforms. We use non-negative matrix factorization to identify user segments and develop personas by adding personality such as names and photos. Our approach can develop accurate personas representing real groups of people using online user data, versus relying on manually gathered data.

Distributed Knowledge Discovery For Diverse Data, Hossein Hamooni

Computer Science ETDs

In the era of new technologies, computer scientists deal with massive data of size hundreds of terabytes. Smart cities, social networks, health care systems, large sensor networks, etc. are constantly generating new data. It is non-trivial to extract knowledge from big datasets because traditional data mining algorithms run impractically on such big datasets. However, distributed systems have come to aid this problem while introducing new challenges in designing scalable algorithms. The transition from traditional algorithms to the ones that can be run on a distributed platform should be done carefully. Researchers should design the modern distributed algorithms based on the …

Querying And Visualization Of Moving Objects Using Constraint Databases, Semere M. Woldemariam

School of Computing: Dissertations, Theses, and Student Research

Good querying and visualization of moving objects and their trajectories is still an open problem. This thesis investigates three types of moving objects. First, projectiles, whose parabolic motion is difficult to represent. Second, moving objects that slide down a slope. The representation of these objects is challenging because of their accelerating motion. Third, the motion of migrating animals. The motion of migrating animals is challenging because it also involves some spatio-temporal interpolation. The thesis shows a solution to these problems using ideas from physics and an implementation in the MLPQ constraint databases system. The MLPQ implementation enables several complex spatio-temporal …

Homogenization Of Plastic Deformation In Heterogeneous Lamella Structures, Rui Yuan, Irene J. Beyerlein, Caizhi Zhou

Materials Science and Engineering Faculty Research & Creative Works

It has been shown that unlike its constituent nanocrystalline (NC) phase, a heterogeneous lamella (HL) composite comprising NC and coarse-grain layers exhibits greatly improved ductility. To understand the origin of this enhancement, we present a 3D discrete dislocation, crystal plasticity finite element model to study the development of strains across this microstructure. Here we show that the HL structure homogenizes the plastic strains in the NC layer, weakening the effect of strain concentrations. These findings can provide valuable insight into the effects of material length scales on material instabilities, which is needed to design heterogeneous structures with superior properties.

A Domain Based Approach To Social Relation Recognition, Qianru Sun, Bernt Schiele, Mario Fritz

Research Collection School Of Computing and Information Systems

Social relations are the foundation of human daily life. Developing techniques to analyze such relations from visual data bears great potential to build machines that better understand us and are capable of interacting with us at a social level. Previous investigations have remained partial due to the overwhelming diversity and complexity of the topic and consequently have only focused on a handful of social relations. In this paper, we argue that the domain-based theory from social psychology is a great starting point to systematically approach this problem. The theory provides coverage of all aspects of social relations and equally is …

Demographics Of News Sharing In The U.S. Twittersphere, Julio C.S. Reis, Haewoon Kwak, Jisun An, Johnnatan Messias, Benevenuto Fabrıcio.

Research Collection School Of Computing and Information Systems

The widespread adoption and dissemination of online news through social media systems have been revolutionizing many segments of our society and ultimately our daily lives. In these systems, users can play a central role as they share content to their friends. Despite that, little is known about news spreaders in social media. In this paper, we provide the first of its kind in-depth characterization of news spreaders in social media. In particular, we investigate their demographics, what kind of content they share, and the audience they reach. Among our main findings, we show that males and white users tend to …

A Nash Equilibrium Formulation Of A Tradable Credits Scheme For Incentivizing Transport Choices: From Next-Generation Public Transport Mode Choice To Hot Lanes, Salem Lahlou, Laura Wynter

Research Collection School Of Computing and Information Systems

We consider a tradable credits scheme for binary transport games where one option is faster (or more comfortable) than the other, but its quality of service suffers when usage is high. Applications can be found in mode choice (public transit versus road transport), premium (i.e., express bus) versus ordinary public transit, and fast (e.g., high-occupancy toll, or HOT) versus regular lanes on expressways. We are motivated in particular by the choice between public transport and use of the road network as a privilege to be discouraged. In a future where GPS-based time-distance-place road charging exists, such next-generation transport management strategies …

A Weighted Maximum Matching Algorithm For Influence Maximization And Structural Controllability, Giorgio Sartor, Yeow Khiang Chia, Laura Wynter, Justin Ruths

Research Collection School Of Computing and Information Systems

Structural control and influence maximization on networks both admit the problem of selecting a particular subset of nodes. In structural control, the subset of nodes should guarantee the controllability of the network (in the usual sense) for almost any combination of weights. In influence maximization, given a diffusion process over the network, the chosen subset of nodes (of a given cardinality) should produce the greatest diffusive influence over the rest of the network. While structural control exploits only the structure of the network, influence maximization depends both on the structure and the weights of the edges. We modify an algorithm …

Design Study Of A Heavy Duty Load Cell Using Finite Element Analysis: A Practical Introduction To Mechatronic Design Process, Mohamad Saleh

The ITB Journal

Mechatronics design process is a series of analytical brain storming operations from sepcification to implementation. The mechtronic products are widly available in the market for various use and applications. This is expected to develop futher in the future with great competitiveness. In order to succeed in the market, mechatronic products need to satisfy the market expectations with regard to quality, fitness for purpose, customer's appeal and cost efficiency. Therefore, the design analysis techniques play a significat part in the market success of these products. Finite Element Analysis is probably the most commonly used numerical technique for mechatronic product design. This …

The R Journal (June 2017) 9(1): Complete Issue, The R Foundation

The R Journal

Editorial, Roger Bivand

Contributed Research Articles

iotools: High-Performance I/O Tools for R, Taylor Arnold, Michael J. Kane, and Simon Urbanek

IsoGeneGUI: Multiple Approaches for Dose-Response Analysis of Microarray Data Using R, Martin Otava, Rudradev Sengupta, Ziv Shkedy, Dan Lin, Setia Pramana, Tobias Verbeke, Philippe Haldermans, Ludwig A. Hothorn, Daniel Gerhard, Rebecca M. Kuiper, Florian Klinglmueller, and Adetayo Kasim

Network Visualization with ggplot2, Sam Tyner, François Briatte, and Heike Hofmann

OrthoPanels: An R Package for Estimating a Dynamic Panel Model with Fixed Effects Using the Orthogonal Reparameterization Approach, Mark Pickup, Paul Gustafson, Davor Cubranic, and Geoffrey Evans

The mosaic Package: Helping …

The Noisefiltersr Package: Label Noise Preprocessing In R, Pablo Morales, Julián Luengo, Luís P.F. Garcia, Ana C. Lorena, André C.P.L.F. De Carvalho

The R Journal

In Data Mining, the value of extracted knowledge is directly related to the quality of the used data. This makes data preprocessing one of the most important steps in the knowledge discovery process. A common problem affecting data quality is the presence of noise. A training set with label noise can reduce the predictive performance of classification learning techniques and increase the overfitting of classification models. In this work we present the NoiseFiltersR package. It contains the first extensive R implementation of classical and state-of-the-art label noise filters, which are the most common techniques for preprocessing label noise. The algorithms …

Hosting Data Packages Via Drat: A Case Study With Hurricane Exposure Data, G Brooke Anderson, Dirk Eddelbuettel

The R Journal

Data-only packages offer a way to provide extended functionality for other R users. However, such packages can be large enough to exceed the package size limit (5 megabytes) for the Comprehensive R Archive Network (CRAN). As an alternative, large data packages can be posted to additional repostiories beyond CRAN itself in a way that allows smaller code packages on CRAN to access and use the data. The drat package facilitates creation and use of such alternative repositories and makes it particularly simple to host them via GitHub. CRAN packages can draw on packages posted to drat repositories through the use …

Milr: Multiple-Instance Logistic Regression With Lasso Penalty, Ping-Yang Chen, Ching-Chuan Chen, Chun-Hao Yang, Sheng-Mao Chang, Kuo-Jung Lee

The R Journal

The purpose of the milr package is to analyze multiple-instance data. Ordinary multiple instance data consists of many independent bags, and each bag is composed of several instances. The statuses of bags and instances are binary. Moreover, the statuses of instances are not observed, whereas the statuses of bags are observed. The functions in this package are applicable for analyzing multiple-instance data, simulating data via logistic regression, and selecting important covariates in the regression model. To this end, maximum likelihood estimation with an expectation-maximization algorithm is implemented for model estimation, and a lasso penalty added to the likelihood function is …

Flan: An R Package For Inference On Mutation Models, Adrien Mazoyer, Rémy Drouilhet, Stéphane Despréaux, Bernard Ycart

The R Journal

This paper describes flan, a package providing tools for fluctuation analysis of mutant cell counts. It includes functions dedicated to the distribution of final numbers of mutant cells. Parametric estimation and hypothesis testing are also implemented, enabling inference on different sorts of data with several possible methods. An overview of the subject is proposed. The general form of mutation models is described, including the classical models as particular cases. Estimating from a model, when the data have been generated by another, induces different possible biases, which are identified and discussed. The three estimation methods available in the package are …

Psf: Introduction To R Package For Pattern Sequence Based Forecasting Algorithm, Neeraj Bokde, Gualberto Asencio-Cortés, Francisco Martínez-Álvarez, Kishore Kulat

The R Journal

This paper introduces the R package that implements the Pattern Sequence based Forecasting (PSF) algorithm, which was developed for univariate time series forecasting. This algorithm has been successfully applied to many different fields. The PSF algorithm consists of two major parts: clustering and prediction. The clustering part includes selection of the optimum number of clusters. It labels time series data with reference to such clusters. The prediction part includes functions like optimum window size selection for specific patterns and prediction of future values with reference to past pattern sequences. The PSF package consists of various functions to implement the PSF …