Computer Sciences Commons^™

Visual Attention Methods In Deep Learning: An In-Depth Survey, Mohammed Hassanin, Anwar Saeed, Ibrahim Radwan, Fahad Shahbaz Khan, Ajmal Mian

Computer Vision Faculty Publications

Inspired by the human cognitive system, attention is a mechanism that imitates the human cognitive awareness about specific information, amplifying critical details to focus more on the essential aspects of data. Deep learning has employed attention to boost performance for many applications. Interestingly, the same attention design can suit processing different data modalities and can easily be incorporated into large networks. Furthermore, multiple complementary attention mechanisms can be incorporated in one network. Hence, attention techniques have become extremely attractive. However, the literature lacks a comprehensive survey specific to attention techniques to guide researchers in employing attention in their deep models. …

Iot Clusters Platform For Data Collection, Analysis, And Visualization Use Case, Soin Abdoul Kassif Baba M Traore

Symposium of Student Scholars

Climate change is happening, and many countries are already facing devastating consequences. Populations worldwide are adapting to the season's unpredictability they relay to lands for agriculture. Our first research was to develop an IoT Clusters Platform for Data Collection, analysis, and visualization. The platform comprises hardware parts with Raspberry Pi and Arduino's clusters connected to multiple sensors. The clusters transmit data collected in real-time to microservices-based servers where the data can be accessed and processed. Our objectives in developing this platform were to create an efficient data collection system, relatively cheap to implement and easy to deploy in any part …

Machine Learning-Oriented Predictive Maintenance (Pdm) Framework For Autonomous Vehicles (Avs): Adopting Blockchain For Pdm Solution, Md Jobair Hossain Faruk, Hossain Shahriar, Maria Valero

Symposium of Student Scholars

Autonomous Vehicles (AVs) refers to smart, connected and multimedia cars with technological megatrends of the fourth industrial revolution (Industry 4.0) and have gained huge strive in today's world. AVs adopt automated driving systems (ADS) technique that permits the vehicle to manage and control driving points without human drivers by utilizing advanced equipment including a combination of sensors, controllers, onboard computers, actuators, algorithms, and advanced software embedded in the different parts of the vehicle. These advanced sensors provide unique inputs to the ADS to generate a path from point A to point B. Ensuring the safety of sensors by limiting maintenance …

Students Certification Management (Scm): Hyperledger Fabric-Based Digital Repository, Md Jobair Hossain Faruk, Hossain Shahriar, Maria Valero

Symposium of Student Scholars

The higher education sector has been heavily impacted financially by the economic downturn caused by the pandemic that has resulted a decline in student enrollments. Finding cost-effective novel technology for storing and sharing student's credentials among academic institutions and potential employers is a demand. Within the current conventional approach, ensuring authentication of a candidate’s credentials is costly and time-consuming which gives burdens to thousands of prospective students and potential employees. As a result, candidates fail to secure opportunities for either delay or non-submission of credentials all over the world. Blockchain technology has the potential for students' control over their credentials; …

Computer Simulation Of Raman Spectra And Mode Assignment: Application To Methane, Oluwaseun Omodemi, Ciara Tyler, Martina Kaledin

Symposium of Student Scholars

This work uses driven molecular dynamics (DMD) method, in conjunction with an analytic PES calculated using MP2/aug-cc-pVDZ energies to identify and assign Raman vibrational modes of methane. Recently, a new linearized approach was proposed for the Polarizability Tensor Surfaces (PTS) that yields a unique solution to the least-squares fitting problem and provides a competitive level of accuracy compared to the non-linear PTS model. We used the previously reported B3LYP/6-31+G(d) molecular geometries for CH₄ and generated a new PTS at the MP2/aug-cc-pVDZ level of theory. The performance of the linearly parametrized functional form for the CH₄ PTS is examined. …

A Review Of Dark Web: Crawling And Discovery Of Information, Timothy Williams, Edwin Matthew, Juanjose Rodriguez-Cardenas, Jack Wright, Hossain Shahriar

Symposium of Student Scholars

The dark web is often discussed in taboo by many who are unfamiliar with the subject. However, this essay takes a dive into the skeleton of what constructs the dark web by compiling the research of published essays. TOR and other discussed browsers are specialized web browsers that provide anonymity by going through multiple servers and encrypted networks between the host and client, hiding the IP address of both ends. This provides difficulty in terms of controlling or monitoring the dark web, leading to its popularity in criminal underworlds.

In this work, we provide an overview of data mining and …

Deep Learning: The Many Approaches Of Intrusion Detection System Can Be Implemented And Improved Upon, Trinity Taylor

Cybersecurity Undergraduate Research Showcase

For my research topic I decided to look at Deep learning. Deep learning can be used in many ways for example in web searching. Deep learning can also can improve new businesses and products. Deep learning could lead to amazing discoveries. Deep learning is making a neural network learn something. In my research I talk about Intrusion detection system, traditional approach for intrusion detection, existing intrusion detection, machine learning and deep learning based intrusion detection system, and future work.

Computer Simulations And Network-Based Profiling Of Binding And Allosteric Interactions Of Sars-Cov-2 Spike Variant Complexes And The Host Receptor: Dissecting The Mechanistic Effects Of The Delta And Omicron Mutations, Gennady M. Verkhivker, Steve Agajanian, Ryan Kassab, Keerthi Krishnan

Mathematics, Physics, and Computer Science Faculty Articles and Research

In this study, we combine all-atom MD simulations and comprehensive mutational scanning of S-RBD complexes with the angiotensin-converting enzyme 2 (ACE2) host receptor in the native form as well as the S-RBD Delta and Omicron variants to (a) examine the differences in the dynamic signatures of the S-RBD complexes and (b) identify the critical binding hotspots and sensitivity of the mutational positions. We also examined the differences in allosteric interactions and communications in the S-RBD complexes for the Delta and Omicron variants. Through the perturbation-based scanning of the allosteric propensities of the SARS-CoV-2 S-RBD residues and dynamics-based network centrality and …

Improving Memory Forensics Capabilities On Apple M1 Computers, Raphaela Santos Mettig Rocha

LSU Master's Theses

Malware threats are rapidly evolving to use more sophisticated attacks. By abusing rich application APIs such as Objective-C’s, they are able to gather information about user activity, launch background processes without the user’s knowledge as well as perform other malicious activities. In some cases, memory forensics is the only way to recover artifacts related to this malicious activity, as is the case with memory-only execution. The introduction of the Rosetta 2 on the Apple M1 introduces a completely new attack surface by allowing binaries of both Intel x86 64 and ARM64 architecture to run in userland. For this reason it …

Deapsecure Computational Training For Cybersecurity: Third-Year Improvements And Impacts, Bahador Dodge, Jacob Strother, Rosby Asiamah, Karina Arcaute, Wirawan Purwanto, Masha Sosonkina, Hongyi Wu

Modeling, Simulation and Visualization Student Capstone Conference

The Data-Enabled Advanced Training Program for Cybersecurity Research and Education (DeapSECURE) was introduced in 2018 as a non-degree training consisting of six modules covering a broad range of cyberinfrastructure techniques, including high performance computing, big data, machine learning and advanced cryptography, aimed at reducing the gap between current cybersecurity curricula and requirements needed for advanced research and industrial projects. By its third year, DeapSECURE, like many other educational endeavors, experienced abrupt changes brought by the COVID-19 pandemic. The training had to be retooled to adapt to fully online delivery. Hands-on activities were reformatted to accommodate self-paced learning. In this paper, …

Applications Of Parallel Discrete Event Simulation, Erik J. Jensen

Modeling, Simulation and Visualization Student Capstone Conference

This work presents three applications of parallel discrete event simulation (PDES), which describe the motivation for and the benefits of using PDES, the kinds of synchronization algorithms that are used, and scaling behavior with these different synchronization algorithms.

Rethinking The Design Of Online Professor Reputation Systems, Haley Tatum

LSU Master's Theses

Online Professor Reputation (OPR) systems, such as RateMyProfessors.com (RMP), are frequently used by college students to post and access peer evaluations of their pro- fessors. However, recent evidence has shown that these platforms suffer from major bias problems. Failing to address bias in online professor ratings not only leads to negative expectations and experiences in class, but also poor performance on exams. To address these concerns, in this thesis, we study bias in OPR systems from a software design point of view. At the first phase of our analysis, we conduct a systematic literature review of 23 interdisciplinary studies on …

A Web User Interface Image Processing Tool For Classifying The Extent Of Dementia Across Alzheimer’S, Sathvik Prasad Palyam, Robin Ghosh

ATU Scholars Symposium

Alzheimer's disease (AD) is the most common form of dementia. This project used four image specifications to classify the dementia stages in each patient applying the CNN algorithm. Employing the CNN-based in silico model, the authors successfully classified and predicted the different AD stages and got around 97.19% accuracy. Later, a web interface tool was developed to educate doctors or researchers to check the patients' dementia level based on the MRI brain images and suggest symptoms that strengthen the predicted level of AI. A user uploads the brain scan, which is sent to the backend server, where the image is …

Fake Profile Detection On Social Media Using Generative Adversarial Networks (Gans), Edidiong Akpan

ATU Scholars Symposium

Generative Adversarial Networks (GANs) is an artificial intelligence framework used to make computers inventive. This has been applied in several areas of security such as intrusion detection systems. With the wave of globalization and increased social media presence, security has become a thing of concern. For instance, social media security breaches and faking has become one of the many things people are subjected to ranging from impersonation, cyberbullying, stalking, fraud, fleecing of personal assets, and issuing online threats behind fake profiles, these fake profiles are created either by hacking an existing account or by copying the lifestyle and events shared …

Internet Of Things Device Capabilities, Architectures, Protocols, And Smart Applications In Healthcare Domain: A Review, Md. Milon Islam, Sheikh Noorduddin, Fakhreddine (Fakhri) Karray, Ghulam Muhammad

Machine Learning Faculty Publications

Nowadays, the Internet has spread to practically every country around the world and is having unprecedented effects on people's lives. The Internet of Things (IoT) is getting more popular and has a high level of interest in both practitioners and academicians in the age of wireless communication due to its diverse applications. The IoT is a technology that enables everyday things to become savvier, everyday computation towards becoming intellectual, and everyday communication to become a little more insightful. In this paper, the most common and popular IoT device capabilities, architectures, and protocols are demonstrated in brief to provide a clear …

Mucot: Multilingual Contrastive Training For Question-Answering In Low-Resource Languages, Gokul Karthik Kumar, Abhishek Singh Gehlot, Sahal Shaji Mullappilly, Karthik Nandakumar

Computer Vision Faculty Publications

Accuracy of English-language Question Answering (QA) systems has improved significantly in recent years with the advent of Transformer-based models (e.g., BERT). These models are pre-trained in a self-supervised fashion with a large English text corpus and further fine-tuned with a massive English QA dataset (e.g., SQuAD). However, QA datasets on such a scale are not available for most of the other languages. Multi-lingual BERT-based models (mBERT) are often used to transfer knowledge from high-resource languages to low-resource languages. Since these models are pre-trained with huge text corpora containing multiple languages, they typically learn language-agnostic embeddings for tokens from different languages. …

Retention Of Qualified Cybersecurity Professionals: A Qualitative Study, Andrew Ishmael, Leila Halawi

Publications

The current endeavors to retain cybersecurity professionals are not enough to sustain the needs of the U.S. workforce. The researchers aim to explore retention strategies for qualified cybersecurity professionals supporting U.S. government contracts at a leading global security company. The researchers interviewed a sample of qualified cybersecurity professionals supporting U.S. government contracts to get essential information on retention in the cybersecurity profession. The researchers proposed four distinct pillars to make up the strategic framework for retaining qualified cybersecurity professionals.

Practical Considerations And Applications For Autonomous Robot Swarms, Rory Alan Hector

LSU Doctoral Dissertations

In recent years, the study of autonomous entities such as unmanned vehicles has begun to revolutionize both military and civilian devices. One important research focus of autonomous entities has been coordination problems for autonomous robot swarms. Traditionally, robot models are used for algorithms that account for the minimum specifications needed to operate the swarm. However, these theoretical models also gloss over important practical details. Some of these details, such as time, have been considered before (as epochs of execution). In this dissertation, we examine these details in the context of several problems and introduce new performance measures to capture practical …

Machine Learning Based Medical Image Deepfake Detection: A Comparative Study, Siddharth Solaiyappan, Yuxin Wen

Engineering Faculty Articles and Research

Deep generative networks in recent years have reinforced the need for caution while consuming various modalities of digital information. One avenue of deepfake creation is aligned with injection and removal of tumors from medical scans. Failure to detect medical deepfakes can lead to large setbacks on hospital resources or even loss of life. This paper attempts to address the detection of such attacks with a structured case study. Specifically, we evaluate eight different machine learning algorithms, which include three conventional machine learning methods (Support Vector Machine, Random Forest, Decision Tree) and five deep learning models (DenseNet121, DenseNet201, ResNet50, ResNet101, VGG19) …

Performing Memory Forensics For Object Recovery From Android Application Memory, Sneha Sudhakaran

LSU Doctoral Dissertations

The analysis of application-specific behavior has become an increasingly important technique in cyber forensics and incident response. The ability to determine the precise actions taken by a user can be the difference between a successful analysis and one that fails to meet its goals. The precise actions includes URLs visited, files downloaded, messages sent and received, images viewed, and data accessed. Evidence extraction from application memory at runtime is an effective solution to successfully extract valuable objects allocated by each application, and it is evident that there is a need for more Android forensics analysis tools that support recovering evidence …

Materialsatlas.Org: A Materials Informatics Web App Platform For Materials Discovery And Survey Of State-Of-The-Art, Jianjun Hu, Stanislav Stefanov, Yuqi Song, Sadman Sadeed Omee, Steph-Yves Louis, Edirisuriya M.D. Siriwardane, Yong Zhao, Lai Wei

Faculty Publications

The availability and easy access of large-scale experimental and computational materials data have enabled the emergence of accelerated development of algorithms and models for materials property prediction, structure prediction, and generative design of materials. However, the lack of user-friendly materials informatics web servers has severely constrained the wide adoption of such tools in the daily practice of materials screening, tinkering, and design space exploration by materials scientists. Herein we first survey current materials informatics web apps and then propose and develop MaterialsAtlas.org, a web-based materials informatics toolbox for materials discovery, which includes a variety of routinely needed tools for exploratory …

Predicting Post-Heart Transplant Composite Renal Outcome Risk In Adults: A Machine Learning Decision Tool, Mutlu Mete, Mehmet U.S. Ayvaci, Venkatesh K. Ariyamuthu, Alpesh Amin

Faculty Publications

No abstract provided.

Cybersecurity And Threat Modeling, Kumar Setty

Computer Science Research Seminars and Symposia

An introduction to cybersecurity and threat modeling.

Multimodal Multi-Head Convolutional Attention With Various Kernel Sizes For Medical Image Super-Resolution, Mariana-Iuliana Georgescu, Radu Tudor Ionescu, Andreea-Iuliana Miron, Olivian Savencu, Nicolae Verga, Nicolae-Cătălin Ristea, Fahad Shabaz Khan

Computer Vision Faculty Publications

Super-resolving medical images can help physicians in providing more accurate diagnostics. In many situations, computed tomography (CT) or magnetic resonance imaging (MRI) techniques output several scans (modes) during a single investigation, which can jointly be used (in a multimodal fashion) to further boost the quality of super-resolution results. To this end, we propose a novel multimodal multi-head convolutional attention module to super-resolve CT and MRI scans. Our attention module uses the convolution operation to perform joint spatial-channel attention on multiple concatenated input tensors, where the kernel (receptive field) size controls the reduction rate of the spatial attention and the number …

Improving Negation Detection With Negation-Focused Pre-Training, Hung Thinh Truong, Timothy Baldwin, Trevor Cohn, Karin Verspoor

Natural Language Processing Faculty Publications

Negation is a common linguistic feature that is crucial in many language understanding tasks, yet it remains a hard problem due to diversity in its expression in different types of text. Recent work has shown that state-of-the-art NLP models underperform on samples containing negation in various tasks, and that negation detection models do not transfer well across domains. We propose a new negation-focused pre-training strategy, involving targeted data augmentation and negation masking, to better incorporate negation information into language models. Extensive experiments on common benchmarks show that our proposed approach improves negation detection performance and generalizability over the strong baseline …

Wrapper And Hybrid Feature Selection Methods Using Metaheuristic Algorithms For English Text Classification: A Systematic Review, Osamah Mohammed Alyasiri, Yu N. Cheah, Ammar Kamal Abasi, Omar Mustafa Al-Janabi

Machine Learning Faculty Publications

Feature selection (FS) constitutes a series of processes used to decide which relevant features/attributes to include and which irrelevant features to exclude for predictive modeling. It is a crucial task that aids machine learning classifiers in reducing error rates, computation time, overfitting, and improving classification accuracy. It has demonstrated its efficacy in myriads of domains, ranging from its use for text classification (TC), text mining, and image recognition. While there are many traditional FS methods, recent research efforts have been devoted to applying metaheuristic algorithms as FS techniques for the TC task. However, there are few literature reviews concerning TC. …

The Death Penalty Is A Viable Punishment, Zane Hirning

Student Research Symposium

The death penalty is a penalty that is worthy of heinous perpetrators. Heinous crimes will be committed, and often, the perpetrator is given a punishment unworthy of the crime committed. This causes a low-risk factor when committing the crime. This can cause issues for many, including the victim’s family. The lack of worthy punishment can affect the victim's family, and it may result in outrage against the public, or a personal attack on the perpetrator or the perpetrator’s family. This creates a violent environment that can lead to other American citizens being hurt because the victim’s family took their own …

Intra-Hour Solar Forecasting Using Cloud Dynamics Features Extracted From Ground-Based Infrared Sky Images, Guillermo Terrén-Serrano

Electrical and Computer Engineering ETDs

Due to the increasing use of photovoltaic systems, power grids are vulnerable to the projection of shadows from moving clouds. An intra-hour solar forecast provides power grids with the capability of automatically controlling the dispatch of energy, reducing the additional cost for a guaranteed, reliable supply of energy (i.e., energy storage). This dissertation introduces a novel sky imager consisting of a long-wave radiometric infrared camera and a visible light camera with a fisheye lens. The imager is mounted on a solar tracker to maintain the Sun in the center of the images throughout the day, reducing the scattering effect produced …

Rotoscoping Image Processing, David J. Blackstone

Student Scholar Showcase

I wrote a program to perform rotoscoping image processing. Rotoscoping is an animation technique that is used to turn real images into cartoon images. This is a lengthy process to do by hand, taking hundreds of hours. To make this job easier, I have written a program to do automated rotoscoping. This program uses many algorithms and techniques such as edge detection, blurring, directional detection, and edge tracking. What once would take hours by hand, takes seconds!

Human-Machine Communication: Complete Volume 4

Human-Machine Communication

This is the complete volume of HMC Volume 4.