Computer Sciences Commons^™

Multispectral Image Analysis Using Convolution Neural Networks, Arun D. Kulkarni

Computer Science Faculty Publications and Presentations

Machine learning (ML) techniques are used often to classify pixels in multispectral images. Recently, there is growing interest in using Convolution Neural Networks (CNNs) for classifying multispectral images. CNNs are preferred because of high performance, advances in hardware such as graphical processing units (GPUs), and availability of several CNN architectures. In CNN, units in the first hidden layer view only a small image window and learn low level features. Deeper layers learn more expressive features by combining low level features. In this paper, we propose a novel approach to classify pixels in a multispectral image using deep convolution neural networks …

Automating Intersection Marking Data Collection And Condition Assessment At Scale With An Artificial Intelligence-Powered System, Kun Xie, Huiming Sun, Xiaomeng Dong, Hong Yang, Hongkai Yu

Civil & Environmental Engineering Faculty Publications

Intersection markings play a vital role in providing road users with guidance and information. The conditions of intersection markings will be gradually degrading due to vehicular traffic, rain, and/or snowplowing. Degraded markings can confuse drivers, leading to increased risk of traffic crashes. Timely obtaining high-quality information of intersection markings lays a foundation for making informed decisions in safety management and maintenance prioritization. However, current labor-intensive and high-cost data collection practices make it very challenging to gather intersection data on a large scale. This paper develops an automated system to intelligently detect intersection markings and to assess their degradation conditions with …

Design Of Robust Blockchain-Envisioned Authenticated Key Management Mechanism For Smart Healthcare Applications, Siddhant Thapiyal, Mohammad Wazid, Devesh Pratap Singh, Ashok Kumar Das, Sachin Shetty

VMASC Publications

The healthcare sector is a very crucial and important sector of any society, and with the evolution of the various deployed technologies, like the Internet of Things (IoT), machine learning and blockchain it has numerous advantages. However, in this section, the data is much more vulnerable than others, because the data is strictly private and confidential, and it requires a highly secured framework for the transmission of data between entities. In this article, we aim to design a blockchain-envisioned authentication and key management mechanism for the IoMT-based smart healthcare applications (in short, we call it SBAKM-HS). We compare the various …

Commentary On Healthcare And Disruptive Innovation, Hilary Finch, Affia Abasi-Amefon, Woosub Jung, Lucas Potter, Xavier-Lewis Palmer

Electrical & Computer Engineering Faculty Publications

Exploits of technology have been an issue in healthcare for many years. Many hospital systems have a problem with “disruptive innovation” when introducing new technology. Disruptive innovation is “an innovation that creates a new market by applying a different set of values, which ultimately overtakes an existing market” (Sensmeier, 2012). Modern healthcare systems are historically slow to accept new technological advancements. This may be because patient-based, provider-based, or industry-wide decisions are tough to implement, giving way to dire consequences. One potential consequence is that healthcare providers may not be able to provide the best possible care to patients. For example, …

An Ioe Blockchain-Based Network Knowledge Management Model For Resilient Disaster Frameworks, Amir Javadpour, Farinaz Sabz Ali Pour, Arun Kumar Sangaiah, Weizhe Zhang, Forough Ja'far, Ashish Singh

Engineering Management & Systems Engineering Faculty Publications

The disaster area is a constantly changing environment, which can make it challenging to distribute supplies effectively. The lack of accurate information about the required goods and potential bottlenecks in the distribution process can be detrimental. The success of a response network is dependent on collaboration, coordination, sovereignty, and equal distribution of relief resources. To facilitate these interactions and improve knowledge of supply chain operations, a reliable and dynamic logistic system is essential. This study proposes the integration of blockchain technology, the Internet of Things (IoT), and the Internet of Everything (IoE) into the disaster management structure. The proposed disaster …

Machine-Learning-Based Vulnerability Detection And Classification In Internet Of Things Device Security, Sarah Bin Hulayyil, Shancang Li, Li Da Xu

Information Technology & Decision Sciences Faculty Publications

Detecting cyber security vulnerabilities in the Internet of Things (IoT) devices before they are exploited is increasingly challenging and is one of the key technologies to protect IoT devices from cyber attacks. This work conducts a comprehensive survey to investigate the methods and tools used in vulnerability detection in IoT environments utilizing machine learning techniques on various datasets, i.e., IoT23. During this study, the common potential vulnerabilities of IoT architectures are analyzed on each layer and the machine learning workflow is described for detecting IoT vulnerabilities. A vulnerability detection and mitigation framework was proposed for machine learning-based vulnerability detection in …

Speculative Futures On Chatgpt And Generative Artificial Intelligence (Ai): A Collective Reflection From The Educational Landscape, Aras Bozkurt, Junhong Xiao, Sarah Lambert, Angelica Pazurek, Helen Crompton, Suzan Koseoglu, Robert Farrow, Melissa Bond, Chrissi Nerantzi, Sarah Honeychurch, Maha Bali, Jon Dron, Kamran Mir, Bonnie Stewart, Eamon Costello, Jon Mason, Christian M. Stracke, Enilda Romero-Hall, Apostolos Koutropoulos, Cathy Mae Toquero, Lenandlar Singh, Ahmed Tlili, Kyungmee Lee, Mark Nichols, Ebba Ossiannilsson, Mark Brown, Valerie Irvine, Juliana Elisa Raffaghelli, Gema Santos-Hermosa, Orna Farrell, Taskeen Adam, Ying Li Thong, Sunagul Sani-Bozkurt, Ramesh C. Sharma, Stefan Hrastinski, Petar Jandrić

Teaching & Learning Faculty Publications

While ChatGPT has recently become very popular, AI has a long history and philosophy. This paper intends to explore the promises and pitfalls of the Generative Pre-trained Transformer (GPT) AI and potentially future technologies by adopting a speculative methodology. Speculative future narratives with a specific focus on educational contexts are provided in an attempt to identify emerging themes and discuss their implications for education in the 21st century. Affordances of (using) AI in Education (AIEd) and possible adverse effects are identified and discussed which emerge from the narratives. It is argued that now is the best of times to define …

Analyzing Pulmonary Abnormality With Superpixel Based Graph Neural Networks In Chest X-Ray, Ronaj Pradhan

Dissertations and Theses

In recent years, the utilization of graph-based deep learning has gained prominence, yet its potential in the realm of medical diagnosis remains relatively unexplored. Convolutional Neural Network (CNN) has achieved state-of-the-art performance in areas such as computer vision, particularly for grid-like data such as images. However, they require a huge dataset to achieve top level of performance and challenge arises when learning from the inherent irregular/unordered nature of physiological data. In this thesis, the research primarily focuses on abnormality screening: classification of Chest X-Ray (CXR) as Tuberculosis positive or negative, using Graph Neural Networks (GNN) that uses Region Adjacency Graphs …

Early Detection Of Sars-Cov-2 Variants Through Dynamic Co-Mutation Network Surveillance, Paul W. Bible, Qiang Huang, Huining Qiu, Yong Huang, Fangfang Zheng, Jing Gu, Jian Sun, Yuantao Hao, Yu Liu

Computer Science Faculty publications

Background: Precise public health and clinical interventions for the COVID-19 pandemic has spurred a global rush on SARS-CoV-2 variant tracking, but current approaches to variant tracking are challenged by the flood of viral genome sequences leading to a loss of timeliness, accuracy, and reliability. Here, we devised a new co-mutation network framework, aiming to tackle these difficulties in variant surveillance.

Methods: To avoid simultaneous input and modeling of the whole large-scale data, we dynamically investigate the nucleotide covarying pattern of weekly sequences. The community detection algorithm is applied to a co-occurring genomic alteration network constructed from mutation corpora of weekly …

A Chromosome-Scale Genome Sequence Of Sudangrass (Sorghum Sudanense) Highlights The Genome Evolution And Regulation Of Dhurrin Biosynthesis, Paul W. Bible, Jieqin Li, Wenmiao Tu, Jian Zheng, Peng Jin, Yanlong Liu, Junli Du, Jiacheng Zheng, Yi-Hong Wang, Qiuwen Zhan

Computer Science Faculty publications

Sudangrass [ Sorghum sudanense (Piper) Stapf] is a hybrid between grain sorghum and its wild relative S. bicolor ssp. verticilliflorum and is grown as a forage crop due to its high biomass production and low dhurrin content compared to sorghum. In this study, we sequenced the sudangrass genome and showed that the assembled genome was 715.95 Mb with 35,243 protein-coding genes. Phylogenetic analysis with whole genome proteomes demonstrated that the sudangrass genome was more similar to US commercial sorghums than to its wild relatives and cultivated sorghums from Africa. We confirmed that at seedling stage, sudangrass accessions contained significantly lower …

Evaluating Ai Sentiment Analysis, Aakriti Shah

Honors Program Theses

This paper presents a comparative analysis of human and AI performance on a sentiment analysis task involving the coding of qualitative data from community program transcripts. The results demonstrate promising but imperfect agreement between two AI models, Claude and Bing, versus three human annotators and one expert annotator using the Community Capitals framework categories. While both models achieved fair alignment with human judgment, confusion patterns emerged involving metaphorical language and text overlapping multiple categories. The findings provide a case study for benchmarking conversational AI systems against human baselines to reveal limitations and target improvements. Key gaps center around distinguishing between …

Exploration Of Robotics Need In The Medical Field And Robotic Arm Operation Via Glove Control, Aditi Vijayvergia

Master’s Theses

This thesis project is an exercise in getting hands-on experience in redesigning and modifying a robotic system. It also involves understanding the current need for robotic applications in hospital settings. To achieve the above, a thorough literature review of the current state of robotics in a hospital setting was conducted. Moreover, a number of interviews with medical care professionals were completed. Three main themes were obtained from the literature review and five main themes were obtained from the interviews which will be presented in this thesis report. The next phase of the project involved redesigning a system that is composed …

Digital Energy Platforms Considering Digital Privacy And Security By Design Principles, Umit Cali, Marthe Fogstad Dynge, Ahmed Idries, Sambeet Mishra, Ivanko Dmytro, Naser Hashemipour, Murat Kuzlu, Aleksandra Mileva (Ed.), Steffen Wendzel (Ed.), Virginia Franqueira (Ed.)

Engineering Technology Faculty Publications

The power system and markets have become increasingly complex, along with efforts to digitalize the energy sector. Accessing flexibility services, in particular, through digital energy platforms, has enabled communication between multiple entities within the energy system and streamlined flexibility market operations. However, digitalizing these vast and complex systems introduces new cybersecurity and privacy concerns, which must be properly addressed during the design of the digital energy platform ecosystems. More specifically, both privacy and cybersecurity measures should be embedded into all phases of the platform design and operation, based on the privacy and security by design principles. In this study, these …

Integration Of Omnet++ Into A Networking Course In An Electrical Engineering Technology Program, Murat Kuzlu, Brian Emmanuel Tamayo, Salih Sarp, Otilia Popescu, Vukica M. Jovanovic

Engineering Technology Faculty Publications

Networking courses are an integral part of electrical engineering technology programs as the majority of electronics in the modern day are required to communicate with each other. They are also getting more attention in manufacturing engineering technology programs because of the development of emerging technologies in Industry 4.0 arena. From laptops, computers, cellphones, modern day vehicles and smart refrigerators, these devices require a certain level of networking in order to communicate with other devices, whether it be locally, or even across the other side of the world. The objective of networking courses in an electrical engineering program is to demonstrate …

Understanding And Assessing Demographic (In)Equity Resulting From Extreme Heat Exposure Due To Lack Of Tree Canopies In Norfolk, Va Using Agent-Based Modeling, Virginia Zamponi, Kevin O'Brien, Erik Jensen, Brandon Feldhaus, Russell Moore, Christopher J. Lynch, Ross Gore

VMASC Publications

Prolonged exposure to extreme heat can result in illness and death. In urban areas of dense concentrations of pavement, buildings, and other surfaces that absorb and retain heat, extreme heat conditions can arise regularly and create harmful environmental exposures for residents daily during certain parts of the year. Tree canopies provide shade and help to cool the environment, making mature trees with large canopies a simple and effective way to reduce urban heat. We develop a demographically representative 1 (agent): 1 (person) agent-based model to understand the extent to which different demographics of residents in Norfolk, VA are equitably shaded …

A Structured Narrative Prompt For Prompting Narratives From Large Language Models: Sentiment Assessment Of Chatgpt-Generated Narratives And Real Tweets, Christopher J. Lynch, Erik J. Jensen, Virginia Zamponi, Kevin O'Brien, Erika Frydenlund, Ross Gore

VMASC Publications

Large language models (LLMs) excel in providing natural language responses that sound authoritative, reflect knowledge of the context area, and can present from a range of varied perspectives. Agent-based models and simulations consist of simulated agents that interact within a simulated environment to explore societal, social, and ethical, among other, problems. Simulated agents generate large volumes of data and discerning useful and relevant content is an onerous task. LLMs can help in communicating agents' perspectives on key life events by providing natural language narratives. However, these narratives should be factual, transparent, and reproducible. Therefore, we present a structured narrative prompt …

Strategic Information Design In Selfish Routing With Quantum Response Travelers, Sainath Sanga, Venkata Sriram Siddhardh Nadendla, Mukund Telukunta, Sajal K. Das

Computer Science Faculty Research & Creative Works

Selfish routing begets inefficiency in multi-agent transportation systems, leading to significant economic losses in our society. Although several powerful techniques (e. g., marginal cost pricing) have been proposed to mitigate price-of-anarchy (a measure of inefficiency), social welfare maximization still remains a huge challenge in selfish routing, especially when travelers deviate from maximizing their own expected utilities. This paper proposes a novel informational intervention to improve the efficiency of selfish routing, especially in the presence of quantal response travelers. Specifically, modeling the interaction between the system and travelers as a Stackelberg game, and develop a novel approximate algorithm, called LoRI (which …

Machine Learning-Based Jet And Event Classification At The Electron-Ion Collider With Applications To Hadron Structure And Spin Physics, Kyle Lee, James Mulligan, Mateusz Płoskoń, Felix Ringer, Feng Yuan

Physics Faculty Publications

We explore machine learning-based jet and event identification at the future Electron-Ion Collider (EIC). We study the effectiveness of machine learning-based classifiers at relatively low EIC energies, focusing on (i) identifying the flavor of the jet and (ii) identifying the underlying hard process of the event. We propose applications of our machine learning-based jet identification in the key research areas at the future EIC and current Relativistic Heavy Ion Collider program, including enhancing constraints on (transverse momentum dependent) parton distribution functions, improving experimental access to transverse spin asymmetries, studying photon structure, and quantifying the modification of hadrons and jets in …

Machine-Assisted Discovery Of Integrable Symplectic Mappings, T. Zolkin, Y. Kharkov, S. Nagaitsev

Physics Faculty Publications

We present a new automated method for finding integrable symplectic maps of the plane. These dynamical systems possess a hidden symmetry associated with an existence of conserved quantities, i.e., integrals of motion. The core idea of the algorithm is based on the knowledge that the evolution of an integrable system in the phase space is restricted to a lower-dimensional submanifold. Limiting ourselves to polygon invariants of motion, we analyze the shape of individual trajectories thus successfully distinguishing integrable motion from chaotic cases. For example, our method rediscovers some of the famous McMillan-Suris integrable mappings and ultradiscrete Painlevé equations. In total, …

Dynamic Predictions Of Thermal Heating And Cooling Of Silicon Wafer, Hitesh Kumar

Master's Projects

Neural Networks are now emerging in every industry. All the industries are trying their best to exploit the benefits of neural networks and deep learning to make predictions or simulate their ongoing process with the use of their generated data. The purpose of this report is to study the heating pattern of a silicon wafer and make predictions using various machine learning techniques. The heating of the silicon wafer involves various factors ranging from number of lamps, wafer properties and points taken in consideration to capture the heating temperature. This process involves dynamic inputs which facilitates the heating of the …

A Data Delivery Mechanism For Disconnected Mobile Applications, Shashank Hegde

Master's Projects

Previous attempts to bring the data of the internet to environments that do not have continuous connectivity to the internet have made use of special hardware which requires additional expenditure on installation. We will develop a software-based infrastructure running on existing Android smartphones to exchange application data between a disconnected user’s phone and corresponding application servers on the internet. The goal of this project is to implement client and server modules for this infrastructure to run on a disconnected phone and the internet respectively. These modules will multiplex application data to be sent into packages and distribute the data present …

Security And Routing In A Disconnected Delay Tolerant Network, Anirudh Kariyatil Chandakara

Master's Projects

Providing internet access in disaster-affected areas where there is little to no internet connectivity is extremely difficult. This paper proposes an architecture that utilizes existing hardware and mobile applications to enable users to access the Internet while maintaining a high level of security. The system comprises a client application, a transport application, and a server running on the cloud. The client combines data from all supported applications into a single bundle, which is encrypted using an end-to-end encryption technique and sent to the transport. The transport physically moves the bundles to a connected area and forwards them to the server. …

Automated Evaluation For Distributed System Assignments, Nimesh Nischal

Master's Projects

A distributed system can exist in numerous states, including many erroneous permutations that could have been addressed in the code. As distributed systems such as cloud computing and microservices gain popularity, involving distributed com- puting assignments is becoming increasingly crucial in Computer Science and related fields. However, designing such systems poses various challenges, such as considering parallel executions, error-inducing edge cases, and interactions with external systems. Typically, distributed assignments require students to implement a system and run multiple instances of the same code to behave as distributed. However, such assign- ments do not encourage students to consider the potential edge …

Deep Learning Neural Machine Translation Conversational Agent, Abhishek Vaid

Master's Projects

Neural Machine Translation (NMT) is a prominent natural language processing technique that is being used to develop conversational AI technology. However, most chatbots do not provide live API features and have list-based scripted responses. Most chatbots are majorly restricted by the training data on which they were trained on and have no knowledge of current events. This research project intends to research and develop an approach to providing live information. We experiment with various techniques in terms of the type of data being used to harness live capabilities. We optimize the hyperparameters that are needed for a Conversational AI agent …

Personalized Tweet Recommendation Using Users’ Image Preferences, Shashwat Avinash Kadam

Master's Projects

In the era of information explosion, the vast amount of data on social media platforms can overwhelm users. Not only does this information explosion contain irrelevant content, but also intentionally fabricated articles and images. As a result, personalized recommendation systems have become increasingly important to help users navigate and make sense of this data. We propose a novel technique to use users’ image preferences to recommend tweets. We extract vital information by analyzing images liked by users and use it to recommend tweets from Twitter. As many images online have no descriptive metadata associated with them, in this framework, we …

Visualizing Classification Errors And Mislabeling In Machine Learning, Vedashree Bhandare

Master's Projects

Deep neural networks have gained popularity and achieved high performance across multiple domains like medical decision-making, autonomous vehicles, decision support systems, etc. Despite this achievement, the internal workings of these models are opaque and are considered as black boxes due to their nested and non-linear structure. This opaque nature of the deep neural networks makes it difficult to interpret the reason behind their output, thus reducing trust and verifiability of the system where these models are applied. This paper explains a systematic approach to identify the clusters with most misclassifications or false label annotations. For this research, we extracted the …

Explainable Ai For Android Malware Detection, Maithili Kulkarni

Master's Projects

Android malware detection based on machine learning (ML) is widely used by the mobile device security community. Machine learning models offer benefits in terms of detection accuracy and efficiency, but it is often difficult to understand how such models make decisions. As a result, popular malware detection strategies remain black box models, which may result in a lack of accountability and trust in the decisions made. The field of explainable artificial intelligence (XAI) attempts to shed light on such black box models. In this research, we apply XAI techniques to ML-based Android malware detection systems. We train classic ML models …

Proof-Of-Stake For Spartangold, Nimesh Ashok Doolani

Master's Projects

Consensus protocols are critical for any blockchain technology, and Proof-of- Stake (PoS) protocols have gained popularity due to their advantages over Proof-of- Work (PoW) protocols in terms of scalability and efficiency. However, existing PoS mechanisms, such as delegated and bonded PoS, suffer from security and usability issues. Pure PoS (PPoS) protocols provide a stronger decentralization and offer a potential solution to these problems. Algorand, a well-known cryptocurrency, employs a PPoS protocol that utilizes a new Byzantine Agreement (BA) mechanism for consensus and Verifiable Random Functions (VRFs) to securely scale the protocol to accommodate many participants, making it possible to handle …

Ubiquitous Application Data Collection In A Disconnected Distributed System, Deepak Munagala

Master's Projects

Despite some incredible advancements in technology, a significant population of the world does not have internet connectivity. These people lack access to crucial information that is easily available to the rest of the world. To solve this problem, we implement a Delay Tolerant Network (DTN) that allows users in disconnected regions access to the internet. This is enabled by collecting all data requests on the users’ phones and passing them to a device that can carry them to a connected region. This device can then collect the necessary information and give it back to the users in the disconnected region. …

Graph Deep Learning Based Hashtag Recommender For Reels On Social Media, Sriya Balineni

Master's Projects

Many businesses, including Facebook, Netflix, and YouTube, rely heavily on a recommendation system. Recommendation systems are algorithms that attempt to provide consumers with relevant suggestions for items such as movies, videos, or reels (microvideos) to watch, hashtags for their posts, songs to listen to, and products to purchase. In many businesses, recommender systems are essential because they can generate enormous amounts of revenue and make the platform stand out when compared to others. Reels are a feature of the social media platforms that enable users to create and share videos of up to sixty seconds in length. Individuals, businesses, and …