Physical Sciences and Mathematics Commons^™

A Hybrid Deep Learning Approach For Crude Oil Price Prediction, Hind Aldabagh, Xianrong Zheng, Ravi Mukkamala

Computer Science Faculty Publications

Crude oil is one of the world’s most important commodities. Its price can affect the global economy, as well as the economies of importing and exporting countries. As a result, forecasting the price of crude oil is essential for investors. However, crude oil price tends to fluctuate considerably during significant world events, such as the COVID-19 pandemic and geopolitical conflicts. In this paper, we propose a deep learning model for forecasting the crude oil price of one-step and multi-step ahead. The model extracts important features that impact crude oil prices and uses them to predict future prices. The prediction model …

Charged Track Reconstruction With Artificial Intelligence For Clas12, Gagik Gavalian, Polykarpos Thomadakis, Angelos Angelopoulos, Nikos Chrisochoides

Computer Science Faculty Publications

In this paper, we present the results of charged particle track reconstruction in CLAS12 using artificial intelligence. In our approach, we use neural networks working together to identify tracks based on the raw signals in the Drift Chambers. A Convolutional Auto-Encoder is used to de-noise raw data by removing the hits that do not satisfy the patterns for tracks, and second Multi-Layer Perceptron is used to identify tracks from combinations of clusters in the drift chambers. Our method increases the tracking efficiency by 50% for multi-particle final states already conducted experiments. The de-noising results indicate that future experiments can run …

Msdrp: A Deep Learning Model Based On Multisource Data For Predicting Drug Response, Haochen Zhao, Xiaoyu Zhang, Qichang Zhao, Yaohang Li, Jianxin Wang

Computer Science Faculty Publications

Motivation: Cancer heterogeneity drastically affects cancer therapeutic outcomes. Predicting drug response in vitro is expected to help formulate personalized therapy regimens. In recent years, several computational models based on machine learning and deep learning have been proposed to predict drug response in vitro. However, most of these methods capture drug features based on a single drug description (e.g. drug structure), without considering the relationships between drugs and biological entities (e.g. target, diseases, and side effects). Moreover, most of these methods collect features separately for drugs and cell lines but fail to consider the pairwise interactions between drugs and cell …

A Structure-Aware Generative Adversarial Network For Bilingual Lexicon Induction, Bocheng Han, Qian Tao, Lusi Li, Zhihao Xiong

Computer Science Faculty Publications

Bilingual lexicon induction (BLI) is the task of inducing word translations with a learned mapping function that aligns monolingual word embedding spaces in two different languages. However, most previous methods treat word embeddings as isolated entities and fail to jointly consider both the intra-space and inter-space topological relations between words. This limitation makes it challenging to align words from embedding spaces with distinct topological structures, especially when the assumption of isomorphism may not hold. To this end, we propose a novel approach called the Structure-Aware Generative Adversarial Network (SA-GAN) model to explicitly capture multiple topological structure information to achieve accurate …

Identifying The Serious Clinical Outcomes Of Adverse Reactions To Drugs By A Multi-Task Deep Learning Framework, Haochen Zhao, Peng Ni, Qichang Zhao, Xiao Liang, Di Ai, Shannon Erhardt, Jun Wang, Yaohang Li, Jiianxin Wang

Computer Science Faculty Publications

Adverse Drug Reactions (ADRs) have a direct impact on human health. As continuous pharmacovigilance and drug monitoring prove to be costly and time-consuming, computational methods have emerged as promising alternatives. However, most existing computational methods primarily focus on predicting whether or not the drug is associated with an adverse reaction and do not consider the core issue of drug benefit-risk assessment-whether the treatment outcome is serious when adverse drug reactions occur. To this end, we categorize serious clinical outcomes caused by adverse reactions to drugs into seven distinct classes and present a deep learning framework, so-called GCAP, for predicting the …

Dfhic: A Dilated Full Convolution Model To Enhance The Resolution Of Hi-C Data, Bin Wang, Kun Liu, Yaohang Li, Jianxin Wang

Computer Science Faculty Publications

Motivation: Hi-C technology has been the most widely used chromosome conformation capture(3C) experiment that measures the frequency of all paired interactions in the entire genome, which is a powerful tool for studying the 3D structure of the genome. The fineness of the constructed genome structure depends on the resolution of Hi-C data. However, due to the fact that high-resolution Hi-C data require deep sequencing and thus high experimental cost, most available Hi-C data are in low-resolution. Hence, it is essential to enhance the quality of Hi-C data by developing the effective computational methods.

Results: In this work, we propose …

Mitigating Anomalous Electricity Consumption In Smart Cities Using An Ai-Based Stacked-Generalization Technique, Arshid Ali, Laiq Khan, Nadeem Javaid, Safdar Hussain Bouk, Abdulaziz Aldegheishem, Nabil Alrahjeh

Computer Science Faculty Publications

Energy management and efficient asset utilization play an important role in the economic development of a country. The electricity produced at the power station faces two types of losses from the generation point to the end user. These losses are technical losses (TL) and non-technical losses (NTL). TLs occurs due to the use of inefficient equipment. While NTLs occur due to the anomalous consumption of electricity by the customers, which happens in many ways; energy theft being one of them. Energy theft majorly happens to cut down on the electricity bills. These losses in the smart grid (SG) are the …

Opioid Use Disorder Prediction Using Machine Learning Of Fmri Data, A. Temtam, Liangsuo Ma, F. Gerard Moeller, M. S. Sadique, K. M. Iftekharuddin, Khan M. Iftekharuddin (Ed.), Weijie Chen (Ed.)

Electrical & Computer Engineering Faculty Publications

According to the Centers for Disease Control and Prevention (CDC) more than 932,000 people in the US have died since 1999 from a drug overdose. Just about 75% of drug overdose deaths in 2020 involved Opioid, which suggests that the US is in an Opioid overdose epidemic. Identifying individuals likely to develop Opioid use disorder (OUD) can help public health in planning effective prevention, intervention, drug overdose and recovery policies. Further, a better understanding of prediction of overdose leading to the neurobiology of OUD may lead to new therapeutics. In recent years, very limited work has been done using statistical …

Gender Classification Via Human Joints Using Convolutional Neural Network, Cheng-En Sung

Master's Projects

With the growing demand for gender-related data on diverse applications, including security systems for ascertaining an individual’s identity for border crossing, as well as marketing purposes of digging the potential customer and tailoring special discounts for them, gender classification has become an essential task within the field of computer vision and deep learning. There has been extensive research conducted on classifying human gender using facial expression, exterior appearance (e.g., hair, clothes), or gait movement. However, within the scope of our research, none have specifically focused gender classification on two-dimensional body joints. Knowing this, we believe that a new prediction pipeline …

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 …

Job Tailored Resume Content Generation, Sumedh Kale

Master's Projects

Generally candidates apply to multiple jobs with a single resume and do not tend to customize their resume to match the job description. This hampers their chances of getting a resume shortlisted for the job. The project aims to help such candidates build job tailored resumes that help them create a customized and targeted resume for a specific job or industry. The tool specifically targets candidates’ employment history, for resume content generation. We then use natural language processing

(NLP) techniques to extract and organize this data into a structured format for the dataset. We experiment with multiple variations of the …

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 …

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 …

Image Captioning Using Reinforcement Learning, Venkat Teja Golamaru

Master's Projects

Image captioning is a crucial technology with numerous applications, including enhancing accessibility for the visually impaired, developing automated image indexing and retrieval systems, and enriching social media experiences. However, accurately describing the content of an image in natural language remains a challenge, particularly in low-resource settings where data and computational power are limited. The most advanced image captioning architectures currently use encoder-decoder structures that incorporate a sequential recurrent prediction model. This study adopts a typical Convolutional Neural Network (CNN) encoder Recurrent Neural Network (RNN) decoder structure for image captioning, but it has framed the problem as a sequential decision-making task. …

Influence Maximization Based On Community Detection And Dominating Sets, Ameya Marathe

Master's Projects

An online platform where various people come together to share information and communicate is called a social network. These platforms are set apart from other means of communication mostly because you can follow and interact also with different people even some you never met, comment on their posts, and re-sharing their posts. Companies such as Amazon and Walmart use these platforms daily for marketing purposes, like spreading information regarding new products and services they offer. They carefully select a subset of users, called influencers, who are usually the ones with high influence over the rest of the users. Influencers receive …

Relationalnet Using Graph Neural Networks For Social Recommendations, Dharahas Tallapally

Master's Projects

Traditional recommender systems create models that can predict user interests based on the user-item relationships. However, these systems often have limited performance due to sparse user behavior data. To address this challenge, researchers are now exploring models for social recommendation that can account for both user- user and user-item relationships based on social networks, and user past behavior, respectively. These models aim to understand each user’s behavior by considering their trusted neighbors and their influence on each other. Specifically, the potential embedding of each user is influenced by their trusted neighbors, who are, in turn, influenced by their own trusted …

Container Caching Optimization Based On Explainable Deep Reinforcement Learning, Divyashree Jayaram

Master's Projects

Serverless edge computing environments use lightweight containers to run IoT services on a need basis i.e only when a service is requested. These containers experience a cold start up latency when starting up. One probable solution to reduce the startup delay is container caching on the edge nodes. Edge nodes are nodes that are closer in proximity to the IoT devices. Efficient container caching strategies are required since the resource availability on these edge devices is limited. Because of this constraint on resources, the container caching strategies should also take proper resource utilization into account. This project tries to further …

Resource Coordination Learning For End-To-End Network Slicing Under Limited State Visibility, Xiang Liu

Master's Projects

This paper discusses a resource coordination problem under limited state visibility to realize end-to-end network slices that are hosted by multiple network domains. We formulate this resource coordination problem as a special type of the multi- armed bandit (MAB) problem called the combinatorial multi-armed bandit (CMAB) problem. Based on this formulation, we convert the problem to a regret minimization problem with a linear objective function and solve it by adapting the Learning with Linear Rewards (LLR) algorithm. In this paper, we present a new hybrid approach that incorporates state reports, which include partial resource information in each domain, into the …

The Bias Report: An Automated News Aggregator For Political Bias Classification And News Summarization, Anant Joshi

Master's Projects

Political polarization is on the rise in the US, driven in large part by divisive news that goes viral on the Internet. Specifically, many media outlets use slanted language and publish misinformation in order to drive user traffic and engagement. Almost 80% of US citizens get their news from online sources, but there is a lack of public safeguards against biased news. A large amount of news is published online every day by media organizations, and it is impossible to manually analyze this amount of data. There is a clear need for automated, public-facing solutions in the current political climate …

Multi-Label Text Classification With Transfer Learning, Likhitha Yelamanchili

Master's Projects

Multi-label text categorization is a crucial task in Natural Language Processing, where each text instance can be simultaneously assigned to numerous labels. This project's goal is to assess how well several deep learning models perform on a real-world dataset for multi-label text classification. We employed data augmentation techniques like Synonym Substitution and Random Word Substitution to address the problem of data imbalance. We conducted experiments on a toxic comment classification dataset to evaluate the effectiveness of several deep learning models including Bi-LSTM, GRU, and Bi-GRU, as well as fine- tuned pre-trained BERT models. Many metrics, including log loss, recall@k, and …

Sign Language Recognition Using A Hybrid Machine Learning Model, Peeyusha Shivayogi

Master's Projects

Sign Language is a visual language used by millions of people around the world. American Sign Language (ASL) is one of the most popular sign languages and the third most popular language in the United States. Automatic recognition of ASL signs can help bridge the communication gap between deaf and hearing individuals. In this project, we explore the use of deep learning models for ASL sign recognition, using the MNIST dataset as a benchmark. We preprocessed the data by reshaping the images to the input layer size of the models and normalized the pixel values. We evaluated five popular deep-learning …

Analyzing Improvement Of Mask R-Cnn On Arms Plates (And Sponges And Coral), James Lee

Master's Projects

Coral Reefs and their diverse array of life forms play a vital role in maintaining the health of our planet's environment. However, due to their fragility, it can be challenging to study the reefs without damaging their delicate ecosystem. To address this issue, researchers have employed non-invasive methods such as using Autonomous Reef Monitoring Structures (ARMS) plates to monitor biodiversity. Data was collected as genetic samples from the plates, and high-resolution photographs were taken. To make the best use of this image data, scientists have turned to machine learning and computer vision. Prior to this study, MASKR-CNN was utilized as …

Video Sign Language Recognition Using Pose Extraction And Deep Learning Models, Shayla Luong

Master's Projects

Sign language recognition (SLR) has long been a studied subject and research field within the Computer Vision domain. Appearance-based and pose-based approaches are two ways to tackle SLR tasks. Various models from traditional to current state-of-the-art including HOG-based features, Convolutional Neural Network, Recurrent Neural Network, Transformer, and Graph Convolutional Network have been utilized to tackle the area of SLR. While classifying alphabet letters in sign language has shown high accuracy rates, recognizing words presents its set of difficulties including the large vocabulary size, the subtleties in body motions and hand orientations, and regional dialects and variations. The emergence of deep …

Navigating Classic Atari Games With Deep Learning, Ayan Abhiranya Singh

Master's Projects

Games for the Atari 2600 console provide great environments for testing reinforcement learning algorithms. In reinforcement learning algorithms, an agent typically learns about its environment via the delivery of periodic rewards. Deep Q-Learning, a variant of Q-Learning, utilizes neural networks which train a Q-function to predict the highest future reward given an input state and action. Deep Q-learning has shown great results in training agents to play Atari 2600 games like Space Invaders and Breakout. However, Deep Q-Learning has historically struggled with learning how to play games with greater emphasis on exploration and delayed rewards, like Ms. PacMan. In this …

Steganographic Capacity Of Selected Machine Learning And Deep Learning Models, Lei Zhang

Master's Projects

As machine learning and deep learning models become ubiquitous, it is inevitable that there will be attempts to exploit such models in various attack scenarios. For example, in a steganographic based attack, information would be hidden in a learning model, which might then be used to gain unauthorized access to a computer, or for other malicious purposes. In this research, we determine the steganographic capacity of various classic machine learning and deep learning models. Specifically, we determine the number of low-order bits of the trained parameters of a given model that can be altered without significantly affecting the performance of …

Leveraging Tweets For Rapid Disaster Response Using Bert-Bilstm-Cnn Model, Satya Pranavi Manthena

Master's Projects

Digital networking sites such as Twitter give a global platform for users to discuss and express their own experiences with others. People frequently use social media to share their daily experiences, local news, and activities with others. Many rescue services and agencies frequently monitor this sort of data to identify crises and limit the danger of loss of life. During a natural catastrophe, many tweets are made in reference to the tragedy, making it a hot topic on Twitter. Tweets containing natural disaster phrases but do not discuss the event itself are not informational and should be labeled as non-disaster …

Keystroke Dynamics And User Identification, Atharva Sharma

Master's Projects

We consider the potential of keystroke dynamics for user identification and authentication. We work with a fixed-text dataset, and focus on clustering users based on the difficulty of distinguishing their typing characteristics. After obtaining a confusion matrix, we cluster users into different levels of classification difficulty based on their typing patterns. Our goal is to create meaningful clusters that enable us to apply appropriate authentication methods to specific user clusters, resulting in an optimized balance between security and efficiency. We use a novel feature engineering method that generates image-like features from keystrokes and employ multiclass Convolutional Neural Networks (CNNs) to …

Defending Ai-Based Automatic Modulation Recognition Models Against Adversarial Attacks, Haolin Tang, Ferhat Ozgur Catak, Murat Kuzlu, Evren Catak, Yanxiao Zhao

Engineering Technology Faculty Publications

Automatic Modulation Recognition (AMR) is one of the critical steps in the signal processing chain of wireless networks, which can significantly improve communication performance. AMR detects the modulation scheme of the received signal without any prior information. Recently, many Artificial Intelligence (AI) based AMR methods have been proposed, inspired by the considerable progress of AI methods in various fields. On the one hand, AI-based AMR methods can outperform traditional methods in terms of accuracy and efficiency. On the other hand, they are susceptible to new types of cyberattacks, such as model poisoning or adversarial attacks. This paper explores the vulnerabilities …

Development Of Machine Learning Based Approach To Predict Fuel Consumption And Maintenance Cost Of Heavy-Duty Vehicles Using Diesel And Alternative Fuels, Sasanka Katreddi

Graduate Theses, Dissertations, and Problem Reports

One of the major contributors of human-made greenhouse gases (GHG) namely carbon dioxide (CO2), methane (CH4), and nitrous oxide (NOX) in the transportation sector and heavy-duty vehicles (HDV) contributing to about 27% of the overall fraction. In addition to the rapid increase in global temperature, airborne pollutants from diesel vehicles also present a risk to human health. Even a small improvement that could potentially drive energy savings to the century-old mature diesel technology could yield a significant impact on minimizing greenhouse gas emissions. With the increasing focus on reducing emissions and operating costs, there is a need for efficient and …

Multimodal Neuron Classification Based On Morphology And Electrophysiology, Aqib Ahmad

Graduate Theses, Dissertations, and Problem Reports

Categorizing neurons into different types to understand neural circuits and ultimately brain function is a major challenge in neuroscience. While electrical properties are critical in defining a neuron, its morphology is equally important. Advancements in single-cell analysis methods have allowed neuroscientists to simultaneously capture multiple data modalities from a neuron. We propose a method to classify neurons using both morphological structure and electrophysiology. Current approaches are based on a limited analysis of morphological features. We propose to use a new graph neural network to learn representations that more comprehensively account for the complexity of the shape of neuronal structures. In …