Computer Sciences Commons^™

Deeprobot: A Hybrid Deep Neural Network Model For Social Bot Detection Based On User Profile Data, Kadhim Hayawi, Sujith Mathew, Neethu Venugopal, Mohammad M. Masud, Pin Han Ho

All Works

Use of online social networks (OSNs) undoubtedly brings the world closer. OSNs like Twitter provide a space for expressing one’s opinions in a public platform. This great potential is misused by the creation of bot accounts, which spread fake news and manipulate opinions. Hence, distinguishing genuine human accounts from bot accounts has become a pressing issue for researchers. In this paper, we propose a framework based on deep learning to classify Twitter accounts as either ‘human’ or ‘bot.’ We use the information from user profile metadata of the Twitter account like description, follower count and tweet count. We name the …

A High Precision Machine Learning-Enabled System For Predicting Idiopathic Ventricular Arrhythmia Origins, Jianwei Zheng, Guohua Fu, Daniele Struppa, Islam Abudayyeh, Tahmeed Contractor, Kyle Anderson, Huimin Chu, Cyril Rakovski

Mathematics, Physics, and Computer Science Faculty Articles and Research

Background: Radiofrequency catheter ablation (CA) is an efficient antiarrhythmic treatment with a class I indication for idiopathic ventricular arrhythmia (IVA), only when drugs are ineffective or have unacceptable side effects. The accurate prediction of the origins of IVA can significantly increase the operation success rate, reduce operation duration and decrease the risk of complications. The present work proposes an artificial intelligence-enabled ECG analysis algorithm to estimate possible origins of idiopathic ventricular arrhythmia at a clinical-grade level accuracy.

Method: A total of 18,612 ECG recordings extracted from 545 patients who underwent successful CA to treat IVA were proportionally sampled into training, …

Robustness Analysis Of Classification Using Recurrent Neural Networks With Perturbed Sequential Input, Guangyi Liu, Arash Amini, Martin Takac, Nader Motee

Machine Learning Faculty Publications

For a given stable recurrent neural network (RNN) that is trained to perform a classification task using sequential inputs, we quantify explicit robustness bounds as a function of trainable weight matrices. The sequential inputs can be perturbed in various ways, e.g., streaming images can be deformed due to robot motion or imperfect camera lens. Using the notion of the Voronoi diagram and Lipschitz properties of stable RNNs, we provide a thorough analysis and characterize the maximum allowable perturbations while guaranteeing the full accuracy of the classification task. We illustrate and validate our theoretical results using a map dataset with clouds …

Developing Reinforcement Learning Algorithms For Robots To Aim And Pour Solid Objects, Haoxuan Li

USF Tampa Graduate Theses and Dissertations

Pouring is one of the most commonly executed tasks in a variety of environments. Thereis less attention paid to pouring solid objects and avoiding spillage. Learning the dynamics for pouring solid objects can be a challenge because the collisions and static frictions between objects make their trajectories less predictable than liquid. Nonetheless, pouring solid objects is an important task in real life. In this work, we propose a solution to help robots aim and pour solid objects. The agents will learn how to interact with the environment and identify the optimal pouring trajectories, then manipulate the arm to aim at …

Pad Beyond The Classroom: Integrating Pad In The Scrum Workplace, Jade S. Weiss

USF Tampa Graduate Theses and Dissertations

Purpose: The “story” format used in Scrum ticket writing is confusing to developers and leadsto insufficient ticket content, which lends to miscommunication between team members and administrators, and disrupts workflow from the bottom up. A burgeoning methodology in Technical Writing, Purpose, Audience, Design (PAD) is an alternative ticket format that is easier to teach developers and improves the aforementioned conditions than the existing “story” format. The goal of this paper is to lay out why and how PAD can benefit developers on smaller Scrum teams who are tasked with writing their own tickets. This paper does not offer solutions for …

Using Intrinsically-Typed Definitional Interpreters To Verify Compiler Optimizations In A Monadic Intermediate Language, Dani Barrack

Dissertations and Theses

Compiler optimizations are critical to the efficiency of modern functional programs. At the same time, optimizations that unintentionally change the semantics of programs can systematically introduce errors into programs that pass through them. The question of how to best verify that optimizations and other program transformations preserve semantics is an important one, given the potential for error introduction. Dependent types allow us to prove that properties about our programs are correct, as well as to design data types and interpreters in such a way that they are correct-by-construction. In this thesis, we explore the use of dependent types and intrinsically-typed …

Extending The Reach Of The Point-To-Set Principle, J. H. Lutz, Neil Lutz, E. Mayordomo

Computer Science Faculty Works

The point-to-set principle of J. Lutz and N. Lutz (2018) has recently enabled the theory of computing to be used to answer open questions about fractal geometry in Euclidean spaces ℝⁿ. These are classical questions, meaning that their statements do not involve computation or related aspects of logic.

In this paper we extend the reach of the point-to-set principle from Euclidean spaces to arbitrary separable metric spaces X. We first extend two fractal dimensions—computability-theoretic versions of classical Hausdorff and packing dimensions that assign dimensions dim(x) and Dim(x) to individual points x ∈ X—to arbitrary separable metric spaces and to arbitrary …

Establishing Trust In Vehicle-To-Vehicle Coordination: A Sensor Fusion Approach, Jakob Veselsky, Jack West, Isaac Ahlgren, George K. Thiruvathukal, Neil Klingensmith, Abhinav Goel, Wenxin Jiang, James C. Davis, Kyuin Lee, Younghyun Kim

Computer Science: Faculty Publications and Other Works

As we add more autonomous and semi-autonomous vehicles (AVs) to our roads, their effects on passenger and pedestrian safety are becoming more important. Despite extensive testing, AVs do not always identify roadway hazards. Failures in object recognition components have already led to several fatal collisions, e.g. as a result of faults in sensors, software, or vantage point. Although a particular AV may fail, there is an untapped pool of information held by other AVs in the vicinity that could be used to identify roadway hazards before they present a safety threat.

Considerations For Radio Frequency Fingerprinting Across Multiple Frequency Channels, Jose A. Gutierrez Del Arroyo, Brett J. Borghetti, Michael A. Temple

Faculty Publications

Radio Frequency Fingerprinting (RFF) is often proposed as an authentication mechanism for wireless device security, but application of existing techniques in multi-channel scenarios is limited because prior models were created and evaluated using bursts from a single frequency channel without considering the effects of multi-channel operation. Our research evaluated the multi-channel performance of four single-channel models with increasing complexity, to include a simple discriminant analysis model and three neural networks. Performance characterization using the multi-class Matthews Correlation Coefficient (MCC) revealed that using frequency channels other than those used to train the models can lead to a deterioration in performance from …

Deep Learning Techniques For Diabetic Retinopathy Classification: A Survey, Mohammad Z. Atwany, Abdulwahab H. Sahyoun, Mohammad Yaqub

Computer Vision Faculty Publications

Diabetic Retinopathy (DR) is a degenerative disease that impacts the eyes and is a consequence of Diabetes mellitus, where high blood glucose levels induce lesions on the eye retina. Diabetic Retinopathy is regarded as the leading cause of blindness for diabetic patients, especially the working-age population in developing nations. Treatment involves sustaining the patient's current grade of vision since the disease is irreversible. Early detection of Diabetic Retinopathy is crucial in order to sustain the patient's vision effectively. The main issue involved with DR detection is that the manual diagnosis process is very time, money, and effort consuming and involves …

Using Temporal Session Types To Analyze Time Complexities Of Concurrent Programs, Joseph M. Walbran

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

Das et al. develop a method for analyzing the time complexity of concurrent, message-passing algorithms. Their method is based on adding timing information to datatypes. Specifically, they use a family of datatypes called session types; these constrain the structure of interactions that may take place over a channel of communication. In Das’s system, the timing properties of an algorithm can be verified by a typechecker: if the timing information in the session types is mismatched, the computer will report a type error. In their paper, Das et al. develop the theory for such a typechecker, but do not provide an …

The Impact Of Dynamic Difficulty Adjustment On Player Experience In Video Games, Chineng Vang

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

Dynamic Difficulty Adjustment (DDA) is a process by which a video game adjusts its level of challenge to match a player’s skill level. Its popularity in the video game industry continues to grow as it has the ability to keep players continuously engaged in a game, a concept referred to as Flow. However, the influence of DDA on games has received mixed responses, specifically that it can enhance player experience as well as hinder it. This paper explores DDA through the Monte Carlo Tree Search algorithm and Reinforcement Learning, gathering feedback from players seeking to understand what about DDA is …

Intrusion Attacks On Automotive Can And Their Detection, Halley M. Paulson

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

The main highway of communication in a vehicle is the Controller Area Network, commonly known by the acronym CAN. Any vulnerability in this network could allow bad actors to block communication between vehicle subsystems, risking the safety of the vehicle’s occupants. With the ever growing list of vulnerabilities being exposed in the CAN, it is critical to address its safety. This paper looks at one of the known vulnerabilities in the data link layer of the CAN and an Intrusion Detection System that could detect attacks on this network. We detail a few processes of the CAN, arbitration and error …

Scheduling Aircraft Departures To Avoid Enroute Congestion, Johannes Martinez

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

When scheduled flights are forecast to overcrowd sections of enroute airspace, an air traffic control authority may need to delay departures. Mixed integer linear programming can be used to compute a schedule that resolves the congestion while bringing the sum of all delays to a minimum. Standard linear programming constraint formulations for such scheduling problems, however, have poor run times for instances of realistic size. A new constraint formulation based on cycles and paths through a route graph reduces run times in computational experiments. It shows particularly strong performance for schedules that approach the worst-case solution times in standard formulations.

Fighting Gerrymandering By Automating Congressional Redistricting, Jacob Jenness

Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal

Gerrymandering is a political problem that the United States has had for more than 200 years. Politicians have taken the dull and routine process of drawing congressional districts and turned it into a highly-partisan process. However, with recent improvements in redistricting algorithms, researchers Harry Levin and Sorelle Friedler have introduced their recursive Divide and Conquer Redistricting Algorithm. This algorithm has the potential to automate the process of congressional redistricting, thereby removing the potential for bias. By utilizing a set of partitioning and swapping algorithms, the Divide and Conquer Redistricting Algorithm achieves desirable goals, such as low population deviation, and high …

Bcse: Blockchain-Based Trusted Service Evaluation Model Over Big Data, Fengyin Li, Xinying Yu, Rui Ge, Yanli Wang, Yang Cui, Huiyu Zhou

Big Data Mining and Analytics

The blockchain, with its key characteristics of decentralization, persistence, anonymity, and auditability, has become a solution to overcome the overdependence and lack of trust for a traditional public key infrastructure on third-party institutions. Because of these characteristics, the blockchain is suitable for solving certain open problems in the service-oriented social network, where the unreliability of submitted reviews of service vendors can cause serious security problems. To solve the unreliability problems of submitted reviews, this paper first proposes a blockchain-based identity authentication scheme and a new trusted service evaluation model by introducing the scheme into a service evaluation model. The new …

Big Data With Cloud Computing: Discussions And Challenges, Amanpreet Kaur Sandhu

Big Data Mining and Analytics

With the recent advancements in computer technologies, the amount of data available is increasing day by day. However, excessive amounts of data create great challenges for users. Meanwhile, cloud computing services provide a powerful environment to store large volumes of data. They eliminate various requirements, such as dedicated space and maintenance of expensive computer hardware and software. Handling big data is a time-consuming task that requires large computational clusters to ensure successful data storage and processing. In this work, the definition, classification, and characteristics of big data are discussed, along with various cloud services, such as Microsoft Azure, Google Cloud, …

Exploiting More Associations Between Slots For Multi-Domain Dialog State Tracking, Hui Bai, Yan Yang, Jie Wang

Big Data Mining and Analytics

Dialog State Tracking (DST) aims to extract the current state from the conversation and plays an important role in dialog systems. Existing methods usually predict the value of each slot independently and do not consider the correlations among slots, which will exacerbate the data sparsity problem because of the increased number of candidate values. In this paper, we propose a multi-domain DST model that integrates slot-relevant information. In particular, certain connections may exist among slots in different domains, and their corresponding values can be obtained through explicit or implicit reasoning. Therefore, we use the graph adjacency matrix to determine the …

Sampling With Prior Knowledge For High-Dimensional Gravitational Wave Data Analysis, He Wang, Zhoujian Cao, Yue Zhou, Zong-Kuan Guo, Zhixiang Ren

Big Data Mining and Analytics

Extracting knowledge from high-dimensional data has been notoriously difficult, primarily due to the so-called "curse of dimensionality" and the complex joint distributions of these dimensions. This is a particularly profound issue for high-dimensional gravitational wave data analysis where one requires to conduct Bayesian inference and estimate joint posterior distributions. In this study, we incorporate prior physical knowledge by sampling from desired interim distributions to develop the training dataset. Accordingly, the more relevant regions of the high-dimensional feature space are covered by additional data points, such that the model can learn the subtle but important details. We adapt the normalizing flow …

Toward Intelligent Financial Advisors For Identifying Potential Clients: A Multitask Perspective, Qixiang Shao, Runlong Yu, Hongke Zhao, Chunli Liu, Mengyi Zhang, Hongmei Song, Qi Liu

Big Data Mining and Analytics

Intelligent Financial Advisors (IFAs) in online financial applications (apps) have brought new life to personal investment by providing appropriate and high-quality portfolios for users. In real-world scenarios, identifying potential clients is a crucial issue for IFAs, i.e., identifying users who are willing to purchase the portfolios. Thus, extracting useful information from various characteristics of users and further predicting their purchase inclination are urgent. However, two critical problems encountered in real practice make this prediction task challenging, i.e., sample selection bias and data sparsity. In this study, we formalize a potential conversion relationship, i.e., user→activated user→client and decompose this relationship into …

A Comparison Of Computational Approaches For Intron Retention Detection, Jiantao Zheng, Cuixiang Lin, Zhenpeng Wu, Hong-Dong Li

Big Data Mining and Analytics

Intron Retention (IR) is an alternative splicing mode through which introns are retained in mature RNAs rather than being spliced in most cases. IR has been gaining increasing attention in recent years because of its recognized association with gene expression regulation and complex diseases. Continuous efforts have been dedicated to the development of IR detection methods. These methods differ in their metrics to quantify retention propensity, performance to detect IR events, functional enrichment of detected IRs, and computational speed. A systematic experimental comparison would be valuable to the selection and use of existing methods. In this work, we conduct an …

Reducing Loading On The Contralateral Limb Using Human-In-The-Loop Optimization, Siena Senatore

UNO Student Research and Creative Activity Fair

In most everyday activities, we head towards a specific goal by updating our choices for a more direct path. However, there are specific clinical tasks where taking the direct path is more challenging. Clinical investigations of optimizing a prosthesis involve the assessment of multiple parameter settings through trial and error rather than goal-directed optimization. We investigate if a human-in-the-loop optimization algorithm can guide manual alterations to a prosthesis-simulating device to reduce the ground reaction force on the contralateral limb. In most participants, the optimal condition reduced the loading rate on the contralateral limb compared to the initial condition tested. These …

A Critical Comparison Of Brave Browser And Google Chrome Forensic Artefacts, Stuart Berham, Sarah Morris

Journal of Digital Forensics, Security and Law

Digital forensic practitioners are tasked with the identification, recovery and analysis of Internet browser artefacts which may have been used in the pursuit of committing a civil or criminal offence. This research paper critically compares the most downloaded browser, Google Chrome, against an increasingly popular Chromium browser known as Brave, said to offer privacy-by-default. With increasing forensic caseloads, data complexity, and requirements for method validation to satisfy ISO 17025 accreditation, recognising the similarities and differences between the browsers, developed on the same underlying technology is essential. The paper describes a series of conducted experiments and subsequent analysis to identify artefacts …

Design And Development Of Software With A Graphical User Interface To Display And Convert Multiple Microscopic Histology Images, Sayed Ahmadreza Razian, Majid Jadidi, Alexey Kamenskiy

UNO Student Research and Creative Activity Fair

Histological images are widely used to assess the microscopic anatomy of biological tissues. Recent advancements in image analysis allow the identification of structural features on histological sections that can help advance medical device development, brain and cancer research, drug discovery, vascular mechanobiology, and many other fields. Histological slide scanners create images in SVS and TIFF formats that were designed to archive image blocks and high-resolution textual information. Because these formats were primarily intended for storage, they are often not compatible with conventional image analysis software and require conversion before they can be used in research. We have developed a user-friendly …

Pseudo-Stereo For Monocular 3d Object Detection In Autonomous Driving, Yi-Nan Chen, Hang Dai, Yong Ding

Computer Vision Faculty Publications

Pseudo-LiDAR 3D detectors have made remarkable progress in monocular 3D detection by enhancing the capability of perceiving depth with depth estimation networks, and using LiDAR-based 3D detection architectures. The advanced stereo 3D detectors can also accurately localize 3D objects. The gap in image-to-image generation for stereo views is much smaller than that in image-to-LiDAR generation. Motivated by this, we propose a Pseudo-Stereo 3D detection framework with three novel virtual view generation methods, including image-level generation, feature-level generation, and feature-clone, for detecting 3D objects from a single image. Our analysis of depth-aware learning shows that the depth loss is effective in …

Three Wave Mixing In Epsilon-Near-Zero Plasmonic Waveguides For Signal Regeneration, Nicholas Mirchandani, Mark C. Harrison

Engineering Faculty Articles and Research

Vast improvements in communications technology are possible if the conversion of digital information from optical to electric and back can be removed. Plasmonic devices offer one solution due to optical computing’s potential for increased bandwidth, which would enable increased throughput and enhanced security. Plasmonic devices have small footprints and interface with electronics easily, but these potential improvements are offset by the large device footprints of conventional signal regeneration schemes, since surface plasmon polaritons (SPPs) are incredibly lossy. As such, there is a need for novel regeneration schemes. The continuous, uniform, and unambiguous digital information encoding method is phase-shift-keying (PSK), so …

The Mathematics Of Risk: An Introduction To Guaranteed Data De-Identification, Kristi Thompson

Western Libraries Presentations

This webinar is devoted to the mathematical and theoretical underpinnings of guaranteed data anonymization. Topics covered include an overview of identifiers and quasi-identifiers, an introduction to k-anonymity, a look at some cases where k-anonymity breaks down, and anonymization hierarchies. The presenter will describe a method to assess a survey dataset for anonymization using standard statistical software and consider the question of "anonymization overkill". Much of the academic material looking at data anonymization is quite abstract and aimed at computer scientists, while material aimed at data curators does not always consider recent developments. This webinar is intended to help bridge the …

An Interactive Game With Virtual Reality Immersion To Improve Cultural Sensitivity In Healthcare, Paul J. Hershberger, Yong Pei, Timothy N. Crawford, Sabrina M. Neeley, Thomas Wischgoll, Dixit B. Patel, Miteshkumar M. Vasoya, Angie Castle, Sankalp Mishra, Lahari Surapaneni, Aman A. Pogaku, Aishwarya Bositty, Todd Pavlack

Computer Science and Engineering Faculty Publications

Purpose: Biased perceptions of individuals who are not part of one’s in-groups tend to be negative and habitual. Because health care professionals are no less susceptible to biases than are others, the adverse impact of biases on marginalized populations in health care warrants continued attention and amelioration. Method: Two characters, a Syrian refugee with limited English proficiency and a black pregnant woman with a history of opioid use disorder, were developed for an online training simulation that includes an interactive life course experience focused on social determinants of health, and a clinical encounter in a community health center utilizing virtual …

Ad-Corre: Adaptive Correlation-Based Loss For Facial Expression Recognition In The Wild, Ali Pourramezan Fard, Mohammad H. Mahoor

Electrical and Computer Engineering: Faculty Scholarship

Automated Facial Expression Recognition (FER) in the wild using deep neural networks is still challenging due to intra-class variations and inter-class similarities in facial images. Deep Metric Learning (DML) is among the widely used methods to deal with these issues by improving the discriminative power of the learned embedded features. This paper proposes an Adaptive Correlation (Ad-Corre) Loss to guide the network towards generating embedded feature vectors with high correlation for within-class samples and less correlation for between-class samples. Ad-Corre consists of 3 components called Feature Discriminator, Mean Discriminator, and Embedding Discriminator. We design the Feature Discriminator component to guide …

A Novel Text Mining Approach For Mental Health Prediction Using Bi-Lstm And Bert Model, Kamil Zeberga, Muhammad Attique, Babar Shah, Farman Ali, Yalew Zelalem Jembre, Tae-Sun Chung

All Works

With the current advancement in the Internet, there has been a growing demand for building intelligent and smart systems that can efficiently address the detection of health-related problems on social media, such as the detection of depression and anxiety. These types of systems, which are mainly dependent on machine learning techniques, must be able to deal with obtaining the semantic and syntactic meaning of texts posted by users on social media. The data generated by users on social media contains unstructured and unpredictable content. Several systems based on machine learning and social media platforms have recently been introduced to identify …