Computer Engineering Commons^™

Efficient Detection Of Diseases By Feature Engineering Approach From Chest Radiograph, Avishek Mukherjee

Legacy Theses & Dissertations (2009 - 2024)

Deep Learning is the new state-of-the-art technology in Image Processing. We applied Deep Learning techniques for identification of diseases from Radiographs made publicly available by NIH. We applied some Feature Engineering approach to augment the data from Anterior-Posterior position to Posterior-Anterior position and vice-versa for all the diseases, at the same point we suppressed ‘No Finding’ radiographs which contributed to more than 50% (approximately 60,000) of the dataset to top 1000 images. We also prepared a model by adding a huge amount of noise to the augmented data, which if need be can be deployed at rural locations which lack …

Emotion Forecasting In Dyadic Conversation : Characterizing And Predicting Future Emotion With Audio-Visual Information Using Deep Learning, Sadat Shahriar

Legacy Theses & Dissertations (2009 - 2024)

Emotion forecasting is the task of predicting the future emotion of a speaker, i.e., the emotion label of the future speaking turn–based on the speaker’s past and current audio-visual cues. Emotion forecasting systems require new problem formulations that differ from traditional emotion recognition systems. In this thesis, we first explore two types of forecasting windows(i.e., analysis windows for which the speaker’s emotion is being forecasted): utterance forecasting and time forecasting. Utterance forecasting is based on speaking turns and forecasts what the speaker’s emotion will be after one, two, or three speaking turns. Time forecasting forecasts what the speaker’s emotion will …

Artificial Intelligence In The Assessment Of Transmission And Distribution Systems Under Natural Disasters Using Machine Learning And Deep Learning Techniques In A Knowledge Discovery Framework, Rossana Villegas

Open Access Theses & Dissertations

Warming trends and increasing temperatures have been observed and reported by federal agencies, such as the National Oceanic and Atmospheric Administration (NOAA). Extreme-weather events, especially hurricanes, tornadoes and winter storms, are among the highly devastating natural disasters responsible for massive and prolonged power outages in Electrical Transmission and Distribution Systems (ETDS). Moreover, the failure rate probability of any system component under extreme-weather tends to increase in the impacted geographic area. This Dissertation proposes an Artificial Intelligence (AI) Decision Support System that can predict damage in the ETDS and allow operators to mitigate disastrous extreme weather events. The document reports the …

Dedicated Hardware For Machine/Deep Learning: Domain Specific Architectures, Angel Izael Solis

Open Access Theses & Dissertations

Artificial intelligence has come a very long way from being a mere spectacle on the silver screen in the 1920s [Hml18]. As artificial intelligence continues to evolve, and we begin to develop more sophisticated Artificial Neural Networks, the need for specialized and more efficient machines (less computational strain while maintaining the same performance results) becomes increasingly evident. Though these “new” techniques, such as Multilayer Perceptron’s, Convolutional Neural Networks and Recurrent Neural Networks, may seem as if they are on the cutting edge of technology, many of these ideas are over 60 years old! However, many of these earlier models, at …

Mediaeval2019: Flood Detection In Time Sequence Satellite Images, Palavi Jain, Bianca Schoen-Phelan, Robert J. Ross

Conference papers

In this work, we present a flood detection technique from time series satellite images for the City-centered satellite sequences (CCSS) task in the MediaEval 2019 competition [1]. This work utilises a three channel feature indexing technique [13] along with a VGG16 pretrained model for automatic detection of floods. We also compared our result with RGB images and a modified NDWI technique by Mishra et al, 2015 [15]. The result shows that the three channel feature indexing technique performed the best with VGG16 and is a promising approach to detect floods from time series satellite images.

Knowledge Graph Reasoning Over Unseen Rdf Data, Bhargavacharan Reddy Kaithi

Browse all Theses and Dissertations

In recent years, the research in deep learning and knowledge engineering has made a wide impact on the data and knowledge representations. The research in knowledge engineering has frequently focused on modeling the high level human cognitive abilities, such as reasoning, making inferences, and validation. Semantic Web Technologies and Deep Learning have an interest in creating intelligent artifacts. Deep learning is a set of machine learning algorithms that attempt to model data representations through many layers of non-linear transformations. Deep learning is in- creasingly employed to analyze various knowledge representations mentioned in Semantic Web and provides better results for Semantic …

Eaglebot: A Chatbot Based Multi-Tier Question Answering System For Retrieving Answers From Heterogeneous Sources Using Bert, Muhammad Rana

College of Graduate Studies: Theses & Dissertations

This paper proposes to tackle Question Answering on a specific domain by developing a multi-tier system using three different types of data storage for storing answers. For testing our system on University domain we have used extracted data from Georgia Southern University website. For the task of faster retrieval we have divided our answer data sources into three distinct types and utilized Dialogflow's Natural Language Understanding engine for route selection. We compared different word and sentence embedding techniques for making a semantic question search engine and BERT sentence embedding gave us the best result and for extracting answer from a …

Investigating Dataset Distinctiveness, Andrew Ulmer, Kent W. Gauen, Yung-Hsiang Lu, Zohar R. Kapach, Daniel P. Merrick

The Summer Undergraduate Research Fellowship (SURF) Symposium

Just as a human might struggle to interpret another human’s handwriting, a computer vision program might fail when asked to perform one task in two different domains. To be more specific, visualize a self-driving car as a human driver who had only ever driven on clear, sunny days, during daylight hours. This driver – the self-driving car – would inevitably face a significant challenge when asked to drive when it is violently raining or foggy during the night, putting the safety of its passengers in danger. An extensive understanding of the data we use to teach computer vision models – …

Deep Neural Network Architectures For Modulation Classification Using Principal Component Analysis, Sharan Ramjee, Shengtai Ju, Diyu Yang, Aly El Gamal

The Summer Undergraduate Research Fellowship (SURF) Symposium

In this work, we investigate the application of Principal Component Analysis to the task of wireless signal modulation recognition using deep neural network architectures. Sampling signals at the Nyquist rate, which is often very high, requires a large amount of energy and space to collect and store the samples. Moreover, the time taken to train neural networks for the task of modulation classification is large due to the large number of samples. These problems can be drastically reduced using Principal Component Analysis, which is a technique that allows us to reduce the dimensionality or number of features of the samples …

Machine Learning Models For Context-Aware Recommender Systems, Yogesh Jhamb

Engineering Ph.D. Theses

The mass adoption of the internet has resulted in the exponential growth of products and services on the world wide web. An individual consumer, faced with this data deluge, is expected to make reasonable choices saving time and money. Organizations are facing increased competition, and they are looking for innovative ways to increase revenue and customer loyalty. A business wants to target the right product or service to an individual consumer, and this drives personalized recommendation. Recommender systems, designed to provide personalized recommendations, initially focused only on the user-item interaction. However, these systems evolved to provide a context-aware recommendations. Context-aware …

Deep Learning And Localized Features Fusion For Medical Image Classification, Haidar A. Almubarak

Doctoral Dissertations

"Local image features play an important role in many classification tasks as translation and rotation do not severely deteriorate the classification process. They have been commonly used for medical image analysis. In medical applications, it is important to get accurate diagnosis/aid results in the fastest time possible.

This dissertation tries to tackle these problems, first by developing a localized feature-based classification system for medical images and using these features and to give a classification for the entire image, and second, by improving the computational complexity of feature analysis to make it viable as a diagnostic aid system in practical clinical …

On Designing An Ecg-Based Intelligent System: Utilizing The Heart’S Electrical Activity To Recognize Humans And Detect Arrhythmia, Sara Saeed Abdeldayem

Graduate Theses, Dissertations, and Problem Reports (ETD)

The electrocardiogram (ECG) signal is the bioelectrical signal that reflects the heart's activity. It has been extensively used as a diagnostic tool since it holds information about the cardiac health condition. However, recent researches have shown that it exhibits an inter-subject variability property. Therefore, it can be used as a biometric-based modality for either identification or verification purposes. Nevertheless, some of the challenges are faced while employing such a signal. For instance, ECG signal is prone to noise, accordingly, noise filters should be designed to remove the noise while keeping the signal properties. Moreover, factors such as medications, health condition, …

Investigating The Application Of Deep Convolutional Neural Networks In Semi-Supervised Video Object Segmentation, Jayadeep Sasikumar

Dissertations

This thesis investigates the different approaches to video object segmentation and the current state-of-the-art in the discipline, focusing on the different deep learning techniques used to solve the problem. The primary contribution of the thesis is the investigation of usefulness of Exponential Linear Units as activation functions for deep convolutional neural architectures trained to perform object semi-supervised segmentation in videos. Mask R-CNN was chosen as the base convolutional neural architecture, with the view of extending the image segmentation algorithm to videos. Two models were created, one with Rectified Linear Units and the other with Exponential Linear Units as the respective …

Learning To Predict Clinical Outcomes From Soft Tissue Sarcoma Mri, Hamidreza Farhidzadeh

USF Tampa Graduate Theses and Dissertations

Soft Tissue Sarcomas (STS) are among the most dangerous diseases, with a 50% mortality rate in the USA in 2016. Heterogeneous responses to the treatments of the same sub-type of STS as well as intra-tumor heterogeneity make the study of biopsies imprecise. Radiologists make efforts to find non-invasive approaches to gather useful and important information regarding characteristics and behaviors of STS tumors, such as aggressiveness and recurrence. Quantitative image analysis is an approach to integrate information extracted using data science, such as data mining and machine learning with biological an clinical data to assist radiologists in making the best recommendation …

Deep Neural Networks As Time Series Forecasters Of Energy Demand, Gregory Merkel

Master's Theses (2009 -)

Short-term load forecasting is important for the day-to-day operation of natural gas utilities. Traditionally, short-term load forecasting of natural gas is done using linear regression, autoregressive integrated moving average models, and artificial neural networks. Many purchasing and operating decisions are made using these forecasts, and there can be high cost to both natural gas utilities and their customers if the short-term load forecast is inaccurate. Therefore, the GasDay lab continues to explore new ways to make better forecasts. Recently, deep neural networks (DNNs) have emerged as a powerful tool in machine learning problems. DNNs have been shown to greatly outperform …

Slither.Io Deep Learning Bot, James Caudill

Computer Engineering

Recent advances in deep learning and computer vision techniques and algorithms have inspired me to create a model application. The game environment used is Slither.io. The system has no previous understanding of the game and is able to learn its surroundings through feature detection and deep learning. Contrary to other agents, my bot is able to dynamically learn and react to its environment. It operates extremely well in early game, with little enemy encounters. It has difficulty transitioning to middle and late game due to limited training time. I will continue to develop this algorithm.

Performance Comparison Of Binarized Neural Network With Convolutional Neural Network, Lopamudra Baruah

Dissertations, Master's Theses and Master's Reports

Deep learning is a trending topic widely studied by researchers due to increase in the abundance of data and getting meaningful results with them. Convolutional Neural Networks (CNN) is one of the most popular architectures used in deep learning. Binarized Neural Network (BNN) is also a neural network which consists of binary weights and activations. Neural Networks has large number of parameters and overfitting is a common problem to these networks. To overcome the overfitting problem, dropout is a solution. Randomly dropping some neurons along with its connections helps to prevent co-adaptations which finally help in reducing overfitting. Many researchers …

Deep Learning Approach For Intrusion Detection System (Ids) In The Internet Of Things (Iot) Network Using Gated Recurrent Neural Networks (Gru), Manoj Kumar Putchala

Browse all Theses and Dissertations

The Internet of Things (IoT) is a complex paradigm where billions of devices are connected to a network. These connected devices form an intelligent system of systems that share the data without human-to-computer or human-to-human interaction. These systems extract meaningful data that can transform human lives, businesses, and the world in significant ways. However, the reality of IoT is prone to countless cyber-attacks in the extremely hostile environment like the internet. The recent hack of 2014 Jeep Cherokee, iStan pacemaker, and a German steel plant are a few notable security breaches. To secure an IoT system, the traditional high-end security …

Investigating The Impact Of Unsupervised Feature-Extraction From Multi-Wavelength Image Data For Photometric Classification Of Stars, Galaxies And Qsos, Annika Lindh

Conference papers

Accurate classification of astronomical objects currently relies on spectroscopic data. Acquiring this data is time-consuming and expensive compared to photometric data. Hence, improving the accuracy of photometric classification could lead to far better coverage and faster classification pipelines. This paper investigates the benefit of using unsupervised feature-extraction from multi-wavelength image data for photometric classification of stars, galaxies and QSOs. An unsupervised Deep Belief Network is used, giving the model a higher level of interpretability thanks to its generative nature and layer-wise training. A Random Forest classifier is used to measure the contribution of the novel features compared to a set …

Investigating The Impact Of Unsupervised Feature-Extraction From Multi-Wavelength Image Data For Photometric Classification Of Stars, Galaxies And Qsos, Annika Lindh

Dissertations

This thesis reviews the current state of photometric classification in Astronomy and identifies two main gaps: a dependence on handcrafted rules, and a lack of interpretability in the more successful classifiers. To address this, Deep Learning and Computer Vision were used to create a more interpretable model, using unsupervised training to reduce human bias.

The main contribution is the investigation into the impact of using unsupervised feature-extraction from multi-wavelength image data for the classification task. The feature-extraction is achieved by implementing an unsupervised Deep Belief Network to extract lower-dimensionality features from the multi-wavelength image data captured by the Sloan Digital …