Computer Sciences Commons^™

Toward Analyzing The Diversity Of Extractive Summaries, Aaron David Hudson

Dissertations and Theses

As the amount of text generated across the internet continues to increase, developing methods for processing that text to glean valuable insights is paramount. Automatic text summarization is one such method that aims to provide a concise and representative summary of input text, allowing users access to the most salient points from a large amount of textual data. However, in working with these summaries, especially those generated from social media data, questions arise about not only the quality of a summary, but also its ability to reflect the diversity of user perspectives. This work examines the quality of summaries with …

Ransomware 2.0: An Emerging Threat To National Security, Mohiuddin Ahmed, Sascha Dominik Dov Bachmann, Abu Barkat Ullah, Shaun Barnett

Research outputs 2022 to 2026

The global Covid-19 pandemic has seen the rapid evolution of our traditional working environment; more people are working from home and the number of online meetings has increased. This trend has also affected the security sector. Consequently, the evolution of ransomware to what is now being described as ‘Ransomware 2.0’ has governments, businesses and individuals alike rushing to secure their data.

Intuitr: A Theorem Prover For Intuitionistic Propositional Logic, Erik Rauer

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

A constructive proof proves the existence of a mathematical object by giving the steps necessary to construct said object. Proofs of this type can be interpreted as an algorithm for creating such an object. Intuitionistic Propositional Logic (IPL) is a propositional logic system wherein all valid proofs are constructive. intuitR is a theorem prover for IPL, that is, it determines whether a given formula is valid in IPL or not. In this paper, we describe how intuitR determines the validity of a formula and review its performance. When compared on a benchmark set of problems, intuitR was determined to solve …

Using Blockchain To Improve Security Of The Internet Of Things, Joshua W. Quist

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

The Internet of Things has increased in popularity in recent years, with daily life now being surrounded by “smart devices.” This network of smart devices, such as thermostats, refrigerators, and even stationary bikes affords us convenience, but at a cost. Security measures are typically inferior on these devices; considering that they collect our data around the clock, this is a big reason for concern. Recent research shows that blockchain technology may be one way to address these security concerns. This paper discusses the Internet of Things and the current issues with how security is handled, discusses how blockchain can shore …

Filling Gaps On The Pareto Front In Multi- And Many-Objective Optimization, Richard Lussier

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

Pareto fronts offer insight into the best found solutions of a given problem. Several algorithms have been developed to help maintain a well-distributed Pareto front and therefore offer a wide variety of solutions. However, in real-world problems, the Pareto front isn’t necessarily a continuous surface and may contain holes and/or discontinuous lines. These irregular areas on the Pareto front are considered gaps. These gaps can either be natural or artificial. In their research, Pellicer, Escudero, Alzueta, and Deb suggest a three-step procedure to find, validate, and fill these gaps. First, they developed an algorithm to generate gap points. Second, they …

Approaches To Broadening Participation With Ap Computer Science Principles, Audrey Le Meur

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

The Advanced Placement Computer Science Principles (AP CSP) course framework was created with the intention of broadening participation in computing. Research has produced mixed results on whether or not the framework succeeds in that goal. Given that teachers have significant freedom in how they choose to teach the AP CSP content, students can have a variety of experiences that may or may not impact their continued participation in CS. In this paper, I compare four different approaches to the AP CSP framework by examining their impact on AP exam scores, self-efficacy and confidence, belongingness and identity, and persistence and interest, …

An Overview Of Redirected Walking Approaches And Techniques In Virtual Reality, Benjamin Goldstein

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

One major obstacle to the ideal of virtual reality is the physical constraints of the user’s location, primarily its limited size. A commonly proposed solution is using redirected walking, defined as manipulation of the user’s experience to alter their walking path, to keep the user within a confined physical space without causing any perceivable sensory distortion for the user. This paper discusses various redirected walking approaches which have been proposed, including predictions of user movement via navigation meshes and simulated users, and subtle redirection techniques using blink-induced change blindness and avatar manipulation.

Bridging The Gap Between Object And Image-Level Representations For Open-Vocabulary Detection, Hanoona Rasheed, Muhammad Maaz, Muhammad Uzair Khattak, Salman Khan, Fahad Shahbaz Khan

Computer Vision Faculty Publications

Existing open-vocabulary object detectors typically enlarge their vocabulary sizes by leveraging different forms of weak supervision. This helps generalize to novel objects at inference. Two popular forms of weak-supervision used in open-vocabulary detection (OVD) include pretrained CLIP model and image-level supervision. We note that both these modes of supervision are not optimally aligned for the detection task: CLIP is trained with image-text pairs and lacks precise localization of objects while the image-level supervision has been used with heuristics that do not accurately specify local object regions. In this work, we propose to address this problem by performing object-centric alignment of …

Unpaired Style Transfer Conditional Generative Adversarial Network For Scanned Document Generation, David Jonathan Hawbaker

Dissertations and Theses

Neural networks are a powerful machine learning tool, especially when trained on a large dataset of relevant high-quality data. Generative adversarial networks, image super resolution and most other image manipulation neural networks require a dataset of images and matching target images for training. Collecting and compiling that data can be time consuming and expensive. This work explores an approach for building a dataset of paired document images with a matching scanned version of each document without physical printers or scanners. A dataset of these document image pairs could be used to train a generative adversarial network or image super resolution …

Region-Adaptive, Error-Controlled Scientific Data Compression Using Multilevel Decomposition, Qian Gong, Ben Whitney, Chengzhu Zhang, Xin Liang, Anand Rangarajan, Jieyang Chen, Lipeng Wan, Paul Ullrich, Qing Liu, Robert Jacob, Sanjay Ranka, Scott Klasky

Computer Science Faculty Research & Creative Works

The increase of computer processing speed is significantly outpacing improvements in network and storage bandwidth, leading to the big data challenge in modern science, where scientific applications can quickly generate much more data than that can be transferred and stored. As a result, big scientific data must be reduced by a few orders of magnitude while the accuracy of the reduced data needs to be guaranteed for further scientific explorations. Moreover, scientists are often interested in some specific spatial/temporal regions in their data, where higher accuracy is required. The locations of the regions requiring high accuracy can sometimes be prescribed …

Wikimarks: Harvesting Relevance Benchmarks From Wikipedia, Laura Dietz, Shubham Chatterjee, Connor Lennox, Sumanta Kashyapi, Pooja Oza, Ben Gamari

Computer Science Faculty Research & Creative Works

We provide a resource for automatically harvesting relevance benchmarks from Wikipedia - which we refer to as "Wikimarks"to differentiate them from manually created benchmarks. Unlike simulated benchmarks, they are based on manual annotations of Wikipedia authors. Studies on the TREC Complex Answer Retrieval track demonstrated that leaderboards under Wikimarks and manually annotated benchmarks are very similar. Because of their availability, Wikimarks can fill an important need for Information Retrieval research. We provide a meta-resource to harvest Wikimarks for several information retrieval tasks across different languages: paragraph retrieval, entity ranking, query-specific clustering, outline prediction, and relevant entity linking and many more. …

Bert-Er: Query-Specific Bert Entity Representations For Entity Ranking, Shubham Chatterjee, Laura Dietz

Computer Science Faculty Research & Creative Works

Entity-oriented search systems often learn vector representations of entities via the introductory paragraph from the Wikipedia page of the entity. As such representations are the same for every query, our hypothesis is that the representations are not ideal for IR tasks. In this work, we present BERT Entity Representations (BERT-ER) which are query-specific vector representations of entities obtained from text that describes how an entity is relevant for a query. Using BERT-ER in a downstream entity ranking system, we achieve a performance improvement of 13-42% (Mean Average Precision) over a system that uses the BERT embedding of the introductory paragraph …

Consemblex: A Consensus-Based Transcriptome Assembly Approach That Extends Consemble And Improves Transcriptome Assembly, Richard Mwaba

School of Computing: Dissertations, Theses, and Student Research

An accurate transcriptome is essential to understanding biological systems enabling omics analyses such as gene expression, gene discovery, and gene-regulatory network construction. However, assembling an accurate transcriptome is challenging, especially for organisms without adequate reference genomes or transcriptomes. While several methods for transcriptome assembly with different approaches exist, it is still difficult to establish the most accurate methods. This thesis explores the different transcriptome assembly methods and compares their performances using simulated benchmark transcriptomes with varying complexity. We also introduce ConSemblEX to improve a consensus-based ensemble transcriptome assembler, ConSemble, in three main areas: we provide the ability to use any …

Circadiomics: Circadian Omic Web Portal, Muntaha Samad, Forest Agostinelli, Tomoki Sato, Kohei Shimaji, Pierre Baldi

Faculty Publications

Circadian rhythms are a foundational aspect of biology. These rhythms are found at the molecular level in every cell of every living organism and they play a fundamental role in homeostasis and a variety of physiological processes. As a result, biomedical research of circadian rhythms continues to expand at a rapid pace. To support this research, CircadiOmics (http://circadiomics.igb.uci.edu/) is the largest annotated repository and analytic web server for high-throughput omic (e.g. transcriptomic, metabolomic, proteomic) circadian time series experimental data. CircadiOmics contains over 290 experiments and over 100 million individual measurements, across >20 unique tissues/organs, and 11 different species. Users are …

Using Magic To Teach Computer Programming, Dale F. Reed, Ronald I. Greenberg

Computer Science: Faculty Publications and Other Works

Magic can be used in project-based instruction to motivate students and provide a meaningful context for learning computer programming. This work describes several magic programs of the “Choose a Number” and “Pick a Card” varieties, making connections to underlying computing concepts.

Magic tricks presented as demonstrations and programming assignments elicit wonder and captivate students’ attention, so that students want to understand and replicate the work to show it to friends and family members. Capturing student interest and curiosity motivates them to learn the underlying programming concepts.

Two “Choose a Number” programs are shown where the computer is able to identify …

Fuzzy Reasoning Procedure For Ontologies Based On Rough Membership Approximation, Armand Florentin Donfack Kana, Babatunde Opeoluwa Akinkunmi

Future Computing and Informatics Journal

One of the major challenges in modeling a real-world domain is how to effectively represent uncertain and incomplete knowledge of that domain. Several techniques for representing uncertainty in ontologies have been proposed with some of the techniques lacking provision for vague inference. The classical tableaux-based algorithm does not provide the flexibility for reasoning over such vague ontologies. However, several extensions of the tableaux-based algorithm have been proposed to cope with fuzzy reasoning. Similarly, several alternative reasoning methods for incomplete, inconsistent, and uncertain ontologies have been proposed. One of the major limitations of most of those techniques is that they require …

Textual Emotion Detection Approaches: A Survey, Mahinda Mahmoud Samy Zidan, Ibrahim Elhenawy, Ahmed R. Abas, Mahmoud Othman

Future Computing and Informatics Journal

Over the past decades, social media attracted individuals to express their feelings on any topic or item, resulting in an incremental growth in the size of created data. These feelings and unstructured data paved the path for business organizations to gather information and build statistical analysis. Various machine learning and natural language processing-based approaches are used for sentiment and emotion analysis. Moreover, deep learning-based approaches recently gained popularity due to their remarkable performance in text analysis. This paper provides a comprehensive overview of the prominent machine learning models applied in emotion analysis. It explores various emotion analysis taxonomies, in addition …

An Empirical Study Towards An Automatic Phishing Attack Detection Using Ensemble Stacking Model, Mahmoud Othman, Hesham Hassan

Future Computing and Informatics Journal

Phishing attacks have become one of the most attacks facing internet users, especially after the COVID-19 pandemic, as most organizations have transferred part or most of their work and communication to become online using well-known tools, like email, Zoom, WebEx, etc. Therefore, cyber phishing attacks have become progressively recent, directly and frankly reflecting the designated website, allowing the attacker to observe everything while the victim is exploring Webpages. Hence, utilizing Artificial Intelligence (AI) techniques has become a necessary approach that could be used to detect such attacks automatically. In this paper, we introduce an empirical analysis for automatic phishing detection …

A Comparative Study On Deep Learning Models For Text Classification Of Unstructured Medical Notes With Various Levels Of Class Imbalance, Hongxia Lu, Louis Ehwerhemuepha, Cyril Rakovski

Mathematics, Physics, and Computer Science Faculty Articles and Research

Background

Discharge medical notes written by physicians contain important information about the health condition of patients. Many deep learning algorithms have been successfully applied to extract important information from unstructured medical notes data that can entail subsequent actionable results in the medical domain. This study aims to explore the model performance of various deep learning algorithms in text classification tasks on medical notes with respect to different disease class imbalance scenarios.

Methods

In this study, we employed seven artificial intelligence models, a CNN (Convolutional Neural Network), a Transformer encoder, a pretrained BERT (Bidirectional Encoder Representations from Transformers), and four typical …

A Monte Carlo Framework For Incremental Improvement Of Simulation Fidelity, Damian Lyons, James Finocchiaro, Misha Novitsky, Chris Korpela

Faculty Publications

Robot software developed in simulation often does not be- have as expected when deployed because the simulation does not sufficiently represent reality - this is sometimes called the `reality gap' problem. We propose a novel algorithm to address the reality gap by injecting real-world experience into the simulation. It is assumed that the robot program (control policy) is developed using simulation, but subsequently deployed on a real system, and that the program includes a performance objective monitor procedure with scalar output. The proposed approach collects simulation and real world observations and builds conditional probability functions. These are used to generate …

Sdq: Stochastic Differentiable Quantization With Mixed Precision, Xijie Huang, Zhiqiang Shen, Shichao Li, Zechun Liu, Xianghong Hu, Jeffry Wicaksana, Eric Xing, Kwang Ting Cheng

Machine Learning Faculty Publications

In order to deploy deep models in a computationally efficient manner, model quantization approaches have been frequently used. In addition, as new hardware that supports mixed bitwidth arithmetic operations, recent research on mixed precision quantization (MPQ) begins to fully leverage the capacity of representation by searching optimized bitwidths for different layers and modules in a network. However, previous studies mainly search the MPQ strategy in a costly scheme using reinforcement learning, neural architecture search, etc., or simply utilize partial prior knowledge for bitwidth assignment, which might be biased on locality of information and is sub-optimal. In this work, we present …

Action-Sufficient State Representation Learning For Control With Structural Constraints, Biwei Huang, Chaochao Lu, Liu Leqi, Josã© Miguel Hernã¡Ndez-Lobato, Clark Glymour, Bernhard Schã¶Lkopf, Kun Zhang

Machine Learning Faculty Publications

Perceived signals in real-world scenarios are usually high-dimensional and noisy, and finding and using their representation that contains essential and sufficient information required by downstream decision-making tasks will help improve computational efficiency and generalization ability in the tasks. In this paper, we focus on partially observable environments and propose to learn a minimal set of state representations that capture sufficient information for decision-making, termed Action-Sufficient state Representations (ASRs). We build a generative environment model for the structural relationships among variables in the system and present a principled way to characterize ASRs based on structural constraints and the goal of maximizing …

Gradient-Free Method For Heavily Constrained Nonconvex Optimization, Wanli Shi, Hongchang Gao, Bin Gu

Machine Learning Faculty Publications

Zeroth-order (ZO) method has been shown to be a powerful method for solving the optimization problem where explicit expression of the gradients is difficult or infeasible to obtain. Recently, due to the practical value of the constrained problems, a lot of ZO Frank-Wolfe or projected ZO methods have been proposed. However, in many applications, we may have a very large number of nonconvex white/black-box constraints, which makes the existing zeroth-order methods extremely inefficient (or even not working) since they need to inquire function value of all the constraints and project the solution to the complicated feasible set. In this paper, …

Identification Of Linear Non-Gaussian Latent Hierarchical Structure, Feng Xie, Biwei Huang, Zhengming Chen, Yangbo He, Zhi Geng, Kun Zhang

Machine Learning Faculty Publications

Traditional causal discovery methods mainly focus on estimating causal relations among measured variables, but in many real-world problems, such as questionnaire-based psychometric studies, measured variables are generated by latent variables that are causally related. Accordingly, this paper investigates the problem of discovering the hidden causal variables and estimating the causal structure, including both the causal relations among latent variables and those between latent and measured variables. We relax the frequently-used measurement assumption and allow the children of latent variables to be latent as well, and hence deal with a specific type of latent hierarchical causal structure. In particular, we define …

How To Detect The Fundamental Frequency: Approach Motivated By Soft Computing And Computational Complexity, Eric Freudenthal, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Psychologists have shown that most information about the mood and attitude of a speaker is carried by the lowest (fundamental) frequency. Because of this frequency's importance, even when the corresponding Fourier component is weak, the human brain reconstruct this frequency based on higher harmonics. The problems is that many people lack this ability. To help them better understand moods and attitudes in social interaction, it is therefore desirable to come up with devices and algorithms that would reconstruct the fundamental frequency. In this paper, we show that ideas from soft computing and computational complexity can be used for this purpose.

Why Time Seems To Pass Slowly For Unpleasant Experiences And Quickly For Pleasant Experiences: An Explanation Based On Decision Theory, Laxman Bokati, Vladik Kreinovich

Departmental Technical Reports (CS)

It is known that our perception of time depends on our level of happiness: time seems to pass slower when we have unpleasant experiences and faster if our experiences are pleasant. Several explanations have been proposed for this effect. However, these explanations are based on specific features of human memory and/or human perception, features that, in turn, need explaining. In this paper, we show that this effect can be explained on a much more basic level of decision theory, without utilizing any specific features of human memory or perception.

Monotonic Bit-Invariant Permutation-Invariant Metrics On The Set Of All Infinite Binary Sequences, Irina Padilla, Vladik Kreinovich

Departmental Technical Reports (CS)

In a computer, all the information about an object is described by a sequence of 0s and 1s. At any given moment of time, we only have partial information, but as we perform more measurements and observations, we get longer and longer sequence that provides a more and more accurate description of the object. In the limit, we get a perfect description by an infinite binary sequence. If the objects are similar, measurement results are similar, so the resulting binary sequences are similar. Thus, to gauge similarity of two objects, a reasonable idea is to define an appropriate metric on …

Physical Trajectories Are Smooth, With Velocities At Least As Continuous As Brownian Motion, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

The fact that the kinetic energy of a particle cannot exceed its overall energy implies that the velocity -- i.e. the derivative of the trajectory -- should be bounded. This means, in effect, that all the trajectories are differentiable (smooth). However, at first glance, there seems to be no direct requirement that the velocities continuously depend on time. In this paper, we show that the properties of electromagnetic field necessitate that the velocities are continuous functions of time -- moreover, that they are at least as continuous as the Brownian motion.

Why Would Anyone Invest In A High-Risk Low-Profit Enterprise?, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Strangely enough, investors invest in high-risk low-profit enterprises as well. At first glance, this seems to contradict common sense and financial basics. However, we show that such investments make perfect sense as long as the related risks are independent from the risks of other investments. Moreover, we show that an optimal investment portfolio should allocate some investment to this enterprise.

Spiking Neural Networks And Their Applications: A Review, Kashu Yamazaki, Viet-Khao Vo-Ho, Darshan Bulsara, Ngan Le

Computer Science and Computer Engineering Faculty Publications and Presentations

The past decade has witnessed the great success of deep neural networks in various domains. However, deep neural networks are very resource-intensive in terms of energy consumption, data requirements, and high computational costs. With the recent increasing need for the autonomy of machines in the real world, e.g., self-driving vehicles, drones, and collaborative robots, exploitation of deep neural networks in those applications has been actively investigated. In those applications, energy and computational efficiencies are especially important because of the need for real-time responses and the limited energy supply. A promising solution to these previously infeasible applications has recently been given …