Open Access. Powered by Scholars. Published by Universities.®

Computer Sciences Commons™

Open Access. Powered by Scholars. Published by Universities.®

Discipline
Institution
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 15211 - 15240 of 63040

Full-Text Articles in Computer Sciences

Automated Reconstructions For The Digital Forensic Examiner Workflow, Ryan P. Montgomery Mar 2022

Automated Reconstructions For The Digital Forensic Examiner Workflow, Ryan P. Montgomery

Theses and Dissertations

One product of a digital forensics examination is a reconstruction of events recorded in the media. A reconstruction places all of the case relevant trace into temporal, identity and associative relationships. Creating this reconstruction is a manual and time consuming process for the examiner. This thesis presents AIER. AIER integrates automation, abstraction and visualization into the Autopsy forensic software to improve the reconstruction process. The integration utilizes a custom Autopsy ingest module to extract and abstract artifact data and an interactive graph-based timeline visualization module. These improvements to the forensic examiner workflow are evaluated through a series of use cases.

Go to article

Performance Of Heterogeneous Multi-Agent Systems With Applications In Combined Arms, Robert J. Wilson Mar 2022

Performance Of Heterogeneous Multi-Agent Systems With Applications In Combined Arms, Robert J. Wilson

Theses and Dissertations

Multi-agent systems show great potential for solving problems in complex and dynamic domains. Such systems comprise multiple individual entities called agents. Agents possessing the same behavior or physical form are called homogeneous while agents which differ in these respects are termed heterogeneous. The overall behavior of the system emerges from the many interactions of its component agents. Most multi-agent systems research to date focuses on systems of homogeneous agents, but recent work suggests that heterogeneous agents may improve system performance in certain tasks. This research examines the impact of heterogeneity on multi-agent system effectiveness and investigates the application of multi-agent …

Go to article

Identifying Characteristics For Success Of Robotic Process Automations, Charles M. Unkrich Mar 2022

Identifying Characteristics For Success Of Robotic Process Automations, Charles M. Unkrich

Theses and Dissertations

In the pursuit of digital transformation, the Air Force creates digital airmen. Digital airmen are robotic process automations designed to eliminate the repetitive high-volume low-cognitive tasks that absorb so much of our Airmen's time. The automation product results in more time to focus on tasks that machines cannot sufficiently perform data analytics and improving the Air Force's informed decision-making. This research investigates the assessment of potential automation cases to ensure that we choose viable tasks for automation and applies multivariate analysis to determine which factors indicate successful projects. The data is insufficient to provide significant insights.

Go to article

Securing Infiniband Networks With End-Point Encryption, Noah B. Diamond Mar 2022

Securing Infiniband Networks With End-Point Encryption, Noah B. Diamond

Theses and Dissertations

The NVIDIA-Mellanox Bluefield-2 is a 100 Gbps high-performance network interface which offers hardware offload and acceleration features that can operate directly on network traffic without routine involvement from the ARM CPU. This allows the ARM multi-core CPU to orchestrate the hardware to perform operations on both Ethernet and RDMA traffic at high rates rather than processing all the traffic directly. A testbed called TNAP was created for performance testing and a MiTM verification process called MiTMVMP is used to ensure proper network configuration. The hardware accelerators of the Bluefield-2 support a throughput of nearly 86 Gbps when using IPsec to …

Go to article

Dds-Cerberus: Improving Security In Dds Middleware Using Kerberos Tickets, Andrew T. Park Mar 2022

Dds-Cerberus: Improving Security In Dds Middleware Using Kerberos Tickets, Andrew T. Park

Theses and Dissertations

The military deploys many IoT in battlefield operations to provide information on terrain and enemy combatants. It also deploys automated robots or UAVs where securing and trusting collected data is essential. Choosing the middleware that handles this message transfer is crucial for real-time operations. Networks with multiple entities, including IoT devices, UAVs, and small computers, require robust middleware facilitating message sending in real-time. Ideally, the middleware would provide QoS to handle lost packets and retransmissions in lossy environments, especially between low-power machines. DDS is a middleware that implements real-time and QoS capabilities by sending messages, not based on endpoints but …

Go to article

Exploring Learning Classifier System Behaviors In Multi-Action, Turn-Based Wargames, Garth J.S. Terlizzi Iii Mar 2022

Exploring Learning Classifier System Behaviors In Multi-Action, Turn-Based Wargames, Garth J.S. Terlizzi Iii

Theses and Dissertations

State of the art game-playing Artificial Intelligence research focuses heavily on non-symbolic learning methods. These methods offer little explainable insight into their decision-making processes. Learning Classifier Systems (LCSs) provide an alternative. LCSs use rule-based learning, guided by a Genetic Algorithm (GA), to produce a human-readable rule-set. This thesis explores LCS usefulness in game-playing agents for multi-agent wargames. Several Multi-Agent Learning Classifier System (MALCS) variants are implemented in the wargame Stratagem MIST: a Zeroeth-Level Classifier System (ZCS), an extended Classifier System (XCS), and an Adaptive Pittsburgh Classifier System (APCS). These algorithms were tested against baseline agents as well as the Online …

Go to article

Incorporating Armed Escorts To The Military Medical Evacuation Dispatching Problem Via Stochastic Optimization And Reinforcement Learning, Andrew G. Gelbard Mar 2022

Incorporating Armed Escorts To The Military Medical Evacuation Dispatching Problem Via Stochastic Optimization And Reinforcement Learning, Andrew G. Gelbard

Theses and Dissertations

The military medical evacuation (MEDEVAC) dispatching problem seeks to determine high-quality dispatching policies to maximize the survivability of casualties within contingency operations. This research leverages applied operations research and machine learning techniques to solve the MEDEVAC dispatching problem and evaluate system performance. More specifically, we develop an infinite-horizon, continuous-time Markov decision process (MDP) model and approximate dynamic programming (ADP) solution approach to generate high-quality policies. The ADP solution approach utilizes an approximate value iteration algorithm strategy incorporating gradient descent Q-learning to approximate the value function. A notional, synthetically-generated scenario in Africa based around the capital city of Niger, Niamey is …

Go to article

An Entity-Component System Based, Ieee Dis Interoperability Interface, Noah W. Scott Mar 2022

An Entity-Component System Based, Ieee Dis Interoperability Interface, Noah W. Scott

Theses and Dissertations

In practice, there are several different methods of organizing data within a given software to fulfil its function. The method known as the Entity-Component System (ECS) is a software architecture where data components define entities. These components are stored as organized lists which are operated upon by systems to inject the system's desired behavior. Data is sent across the networks to communicate between simulation nodes as Protocol Data Units (PDUs). When sending PDUs across a network protocol, each simulation represents a common understanding of the world at the desired level of detail. DIS-compliant simulations are commonly written using an Object-Oriented …

Go to article

Bayesian Convolutional Neural Network With Prediction Smoothing And Adversarial Class Thresholds, Noah M. Miller Mar 2022

Bayesian Convolutional Neural Network With Prediction Smoothing And Adversarial Class Thresholds, Noah M. Miller

Theses and Dissertations

Using convolutional neural networks (CNNs) for image classification for each frame in a video is a very common technique. Unfortunately, CNNs are very brittle and have a tendency to be over confident in their predictions. This can lead to what we will refer to as “flickering,” which is when the predictions between frames jump back and forth between classes. In this paper, new methods are proposed to combat these shortcomings. This paper utilizes a Bayesian CNN which allows for a distribution of outputs on each data point instead of just a point estimate. These distributions are then smoothed over multiple …

Go to article

Using Generative Adversarial Networks To Augment Unmanned Aerial Vehicle Image Classification Training Sets, Benjamin J. Mccloskey Mar 2022

Using Generative Adversarial Networks To Augment Unmanned Aerial Vehicle Image Classification Training Sets, Benjamin J. Mccloskey

Theses and Dissertations

A challenging task in computer vision is finding techniques to improve the object detection and classification capabilities of ML models used for processing images acquired by moving aerial platforms. This research explores if GAN augmented UAV training sets can increase the generalizability of a detection model trained on said data. To answer this question, the YOLOv4-Tiny Object Detection Model was trained with aerial image training sets depicting rural environments. The salient objects within the frames were recreated using various GAN architectures, placed back into the original frames, and the augmented frames appended to the original training sets. GAN augmentation on …

Go to article

Team Air Combat Using Model-Based Reinforcement Learning, David A. Mottice Mar 2022

Team Air Combat Using Model-Based Reinforcement Learning, David A. Mottice

Theses and Dissertations

We formulate the first generalized air combat maneuvering problem (ACMP), called the MvN ACMP, wherein M friendly AUCAVs engage against N enemy AUCAVs, developing a Markov decision process (MDP) model to control the team of M Blue AUCAVs. The MDP model leverages a 5-degree-of-freedom aircraft state transition model and formulates a directed energy weapon capability. Instead, a model-based reinforcement learning approach is adopted wherein an approximate policy iteration algorithmic strategy is implemented to attain high-quality approximate policies relative to a high performing benchmark policy. The ADP algorithm utilizes a multi-layer neural network for the value function approximation regression mechanism. One-versus-one …

Go to article

The Impact Of Visual Feedback And Control Configuration On Pilot-Aircraft Interface Using Head Tracking Technology, Christopher M. Arnold Mar 2022

The Impact Of Visual Feedback And Control Configuration On Pilot-Aircraft Interface Using Head Tracking Technology, Christopher M. Arnold

Theses and Dissertations

Traditional control mechanisms restrict human input on the displays in 5th generation aircraft. This research explored methods for enhancing pilot interaction with large, information dense cockpit displays; specifically, the effects of visual feedback and control button configuration when augmenting cursor control with head tracking technology. Previous studies demonstrated that head tracking can be combined with traditional cursor control to decrease selection times but can increase pilot mental and physical workload. A human subject experiment was performed to evaluate two control button configurations and three visual feedback conditions. A Fitts Law analysis was performed to create predictive models of selection time …

Go to article

A Thematic And Reference Analysis Of Touchless Technologies, Eric R. Curia Mar 2022

A Thematic And Reference Analysis Of Touchless Technologies, Eric R. Curia

Theses and Dissertations

The purpose of this research is to explore the utility and current state of touchless technologies. Five categories of technologies are identified as a result of collecting and reviewing literature: facial/biometric recognition, gesture recognition, touchless sensing, personal devices, and voice recognition. A thematic analysis was conducted to evaluate the advantages and disadvantages of the five categories. A reference analysis was also conducted to determine the similarities between articles in each category. Touchless sensing showed to have the most advantages and least similar references. Gesture recognition was the opposite. Comparing analyses shows more reliable technology types are more beneficial and diverse.

Go to article

The Application Of Virtual Reality In Firefighting Training, Dylan A. Gagnon Mar 2022

The Application Of Virtual Reality In Firefighting Training, Dylan A. Gagnon

Theses and Dissertations

Immersive simulations such as virtual reality is becoming more prevalent for use in training environments for many professions. United States Air Force firefighters may benefit from incorporating VR technology into their training program to increase organizational commitment, job satisfaction, self-efficacy, and job performance. With implementing a new training platform, it is also important to understand the relationship between these variables and the perceived benefits and efficacy of the VR training, which has not yet been studied in previous research. This study addresses this issue by gathering data from fire departments currently fielding a VR fire training platform.

Go to article

Monocular Pose Estimation For Automated Aerial Refueling Via Perspective-N-Point, James C. Lynch Mar 2022

Monocular Pose Estimation For Automated Aerial Refueling Via Perspective-N-Point, James C. Lynch

Theses and Dissertations

Any Automated Aerial Refueling (AAR) solution requires the quick and precise estimation of the relative position and rotation of the two aircraft involved. This is currently accomplished using stereo vision techniques augmented by Iterative Closest Point (ICP), but requires post-processing to account for environmental factors such as boom occlusion. This paper proposes a monocular solution, combining a custom-trained single-shot object detection Convolutional Neural Network (CNN) and Perspective-n-Point (PnP) estimation to calculate a pose estimate with a single image. This solution is capable of pose estimation at contact point (22m) within 7cm of error and a rate of 10Hz, regardless of …

Go to article

Wifitrace: Network-Based Contact Tracing For Infectious Diseases Using Passive Wifi Sensing, Amee Trivedi, Camellia Zakaria, Rajesh Krishna Balan, Ann Becker, George Corey, Prashant Shenoy Mar 2022

Wifitrace: Network-Based Contact Tracing For Infectious Diseases Using Passive Wifi Sensing, Amee Trivedi, Camellia Zakaria, Rajesh Krishna Balan, Ann Becker, George Corey, Prashant Shenoy

Research Collection School Of Computing and Information Systems

Contact tracing is a well-established and effective approach for the containment of the spread of infectious diseases. While Bluetooth-based contact tracing method using phones has become popular recently, these approaches suffer from the need for a critical mass adoption to be effective. In this paper, we present WiFiTrace, a network-centric approach for contact tracing that relies on passive WiFi sensing with no client-side involvement. Our approach exploits WiFi network logs gathered by enterprise networks for performance and security monitoring, and utilizes them for reconstructing device trajectories for contact tracing. Our approach is specifically designed to enhance the efficacy of traditional …

Go to article

Delaunay Walk For Fast Nearest Neighbor: Accelerating Correspondence Matching For Icp, James D. Anderson, Ryan M. Raettig, Josh Larson, Scott L. Nykl, Clark N. Taylor, Thomas Wischgoll Mar 2022

Delaunay Walk For Fast Nearest Neighbor: Accelerating Correspondence Matching For Icp, James D. Anderson, Ryan M. Raettig, Josh Larson, Scott L. Nykl, Clark N. Taylor, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

Point set registration algorithms such as Iterative Closest Point (ICP) are commonly utilized in time-constrained environments like robotics. Finding the nearest neighbor of a point in a reference 3D point set is a common operation in ICP and frequently consumes at least 90% of the computation time. We introduce a novel approach to performing the distance-based nearest neighbor step based on Delaunay triangulation. This greedy algorithm finds the nearest neighbor of a query point by traversing the edges of the Delaunay triangulation created from a reference 3D point set. Our work integrates the Delaunay traversal into the correspondences search of …

Go to article

Microwave Hyperthermia Application With Bioimplant Single Slot Coaxial Antennadesign For Mouse Breast Cancer Treatment, Ahmet Ri̇fat Görgün, Cem Baytöre, Selçuk Çömlekçi̇, Mehmet İbrahi̇m Tuğlu, Adnan Kaya Mar 2022

Microwave Hyperthermia Application With Bioimplant Single Slot Coaxial Antennadesign For Mouse Breast Cancer Treatment, Ahmet Ri̇fat Görgün, Cem Baytöre, Selçuk Çömlekçi̇, Mehmet İbrahi̇m Tuğlu, Adnan Kaya

Turkish Journal of Electrical Engineering and Computer Sciences

In this study, a novel animal model for the breast cancer treatment which contains hyperthermia is proposed. For this main purpose a low cost, interstitial, bioimplant antenna by short ended single slot design is proposed to heat the cancerous tissues. Both the theoretical background of the proposed system and the simulation and measurement results of antenna design are presented. An artificial tissue phantom model has been created under laboratory conditions and then the utility of the proposed antenna has been tested on this model. Artificial tissues have been heated by 25W to (41-44 ? C) in a short time like …

Go to article

Critical Speed Calculation Of A Refurbishment Of 11mw Hydro Power Plant Unit, Ahmet Seli̇m Pehli̇van, Dario Kraljevic, Ivan Triplat, Beste Bahçeci̇ Mar 2022

Critical Speed Calculation Of A Refurbishment Of 11mw Hydro Power Plant Unit, Ahmet Seli̇m Pehli̇van, Dario Kraljevic, Ivan Triplat, Beste Bahçeci̇

Turkish Journal of Electrical Engineering and Computer Sciences

Hydro generator design is a significant issue in terms of safety, efficiency, and energy production sustainability. One of the most crucial issues about design criteria is to satisfy the needs of the project?s critical speed. In this work, the critical speed calculation of an 11 MW hydro power plant was investigated with several design steps. Numerical solution methodologies were implemented using ARMD?. A generator design was developed, and the implementation of the rotor was carried out in Antalya, Turkey. Vibration and displacements of the hydro generator are adequate. The generator unit is active for 3 years and have never encountered …

Go to article

Identification And Mitigation Of Non-Line-Of-Sight Path Effect Using Repeater Forhybrid Ultra-Wideband Positioning And Networking System, Gwo Chin Chung, Mohd. Aqmal Syafiq Kamarudin, It Ee Lee, Soo Fun Tan Mar 2022

Identification And Mitigation Of Non-Line-Of-Sight Path Effect Using Repeater Forhybrid Ultra-Wideband Positioning And Networking System, Gwo Chin Chung, Mohd. Aqmal Syafiq Kamarudin, It Ee Lee, Soo Fun Tan

Turkish Journal of Electrical Engineering and Computer Sciences

At least two decades ago, various applications have been proposed for the implementation of ultra-wideband (UWB) technology, but only a few of them are being realised such as radar detection, home networking, and indoor positioning. Although UWB positioning offers precise locality tracking, the accuracy of the estimation is greatly affected by the non-line-of-sight (NLOS) path effect. In this paper, we propose a hybrid indoor UWB positioning and networking system that utilises the existing repeater of the data network to eliminate the NLOS paths. A switching algorithm is written to identify the existence of NLOS paths based on received signal strength …

Go to article

A Factor Graph Optimization Mapping Based On Normaldistributions Transform, Kedi Zhong, Yuansheng Liu, Jiansuo Yang, Ming Lu, Jun Zhang Mar 2022

A Factor Graph Optimization Mapping Based On Normaldistributions Transform, Kedi Zhong, Yuansheng Liu, Jiansuo Yang, Ming Lu, Jun Zhang

Turkish Journal of Electrical Engineering and Computer Sciences

This paper aims to achieve highly accurate mapping results and real time pose estimation of autonomous vehicle by using the normal distribution transform (NDT) algoritm. A factor graph optimization algorithm (FGO-NDT) is proposed to address the poor real-time performance and pose drift errors of the NDT algorithm. Smooth point cloud data are obtained by multisensor calibration and data preprocessing. NDT registration is then used for lidar odometry and feature matching. The global navigation satellite system (GNSS) data and loop detection results are added to the factor graph framework as the pose constraint factors to optimize the pose trajectory and eliminate …

Go to article

Broadband Internet Access: Equal Access And Opportunity For All, Erin Kotten, Emily Louks, Matt Frank, Marissa Mammenga Mar 2022

Broadband Internet Access: Equal Access And Opportunity For All, Erin Kotten, Emily Louks, Matt Frank, Marissa Mammenga

Master of Social Work Student Policy Advocacy Briefs

The internet has become the primary way households gain access to daily services, including education, healthcare, employment resources, and transportation. In 2016, the United Nations General Assembly designated internet access a basic human right. Millions of Americans are living in the digital divide. Low-wage workers, people of color, children, older adults, individuals with disabilities, the less educated, rural residents, and limited English-speaking households are among those most affected by lack of broadband access in Minnesota. Those without access experience social and economic inequalities. The COVID-19 pandemic heightened the need for broadband access. Minnesota must end the digital divide that limits …

Go to article

Perceptions Of Violations By Artificial And Human Actors Across Moral Foundations, Timothy Maninger, Daniel Burton Shank Mar 2022

Perceptions Of Violations By Artificial And Human Actors Across Moral Foundations, Timothy Maninger, Daniel Burton Shank

Psychological Science Faculty Research & Creative Works

Artificial agents such as robots, chatbots, and artificial intelligence systems can be the perpetrators of a range of moral violations traditionally limited to human actors. This paper explores how people perceive the same moral violations differently for artificial agent and human perpetrators by addressing three research questions: How wrong are moral foundation violations by artificial agents compared to human perpetrators? Which moral foundations do artificial agents violate compared to human perpetrators? What leads to increased blame for moral foundation violations by artificial agents compared to human perpetrators? We adapt 18 human-perpetrated moral violation scenarios that differ by the moral foundation …

Go to article

Machine Learning Methods For Generating High Dimensional Discrete Datasets, Giuseppe Manco, Ettore Ritacco, Antonino Rullo, Domenico Saccà, Edoardo Serra Mar 2022

Machine Learning Methods For Generating High Dimensional Discrete Datasets, Giuseppe Manco, Ettore Ritacco, Antonino Rullo, Domenico Saccà, Edoardo Serra

Computer Science Faculty Publications and Presentations

The development of platforms and techniques for emerging Big Data and Machine Learning applications requires the availability of real-life datasets. A possible solution is to synthesize datasets that reflect patterns of real ones using a two-step approach: first, a real dataset X is analyzed to derive relevant patterns Z and, then, to use such patterns for reconstructing a new dataset X' that preserves the main characteristics of X. This survey explores two possible approaches: (1) Constraint-based generation and (2) probabilistic generative modeling. The former is devised using inverse mining (IFM) techniques, and consists of generating a dataset satisfying given …

Go to article

Removing The Veil: Shining Light On The Lack Of Inclusivity In Cybersecurity Education For Students With Disabilities, Felicia Hellems, Sajal Bhatia Mar 2022

Removing The Veil: Shining Light On The Lack Of Inclusivity In Cybersecurity Education For Students With Disabilities, Felicia Hellems, Sajal Bhatia

School of Computer Science & Engineering Faculty Publications

There are currently over one billion people living with some form of disability worldwide. The continuous increase in new technologies in today's society comes with an increased risk in security. A fundamental knowledge of cybersecurity should be a basic right available to all users of technology. A review of literature in the fields of cybersecurity, STEM, and computer science (CS) has revealed existent gaps regarding educational methods for teaching cybersecurity to students with disabilities (SWD's). To date, SWD's are largely left without equitable access to cybersecurity education. Our goal is to identify current educational methods being used to teach SWD's …

Go to article

Sample-Efficient Iterative Lower Bound Optimization Of Deep Reactive Policies For Planning In Continuous Mdps, Siow Meng Low, Akshat Kumar, Scott Sanner Mar 2022

Sample-Efficient Iterative Lower Bound Optimization Of Deep Reactive Policies For Planning In Continuous Mdps, Siow Meng Low, Akshat Kumar, Scott Sanner

Research Collection School Of Computing and Information Systems

Recent advances in deep learning have enabled optimization of deep reactive policies (DRPs) for continuous MDP planning by encoding a parametric policy as a deep neural network and exploiting automatic differentiation in an end-toend model-based gradient descent framework. This approach has proven effective for optimizing DRPs in nonlinear continuous MDPs, but it requires a large number of sampled trajectories to learn effectively and can suffer from high variance in solution quality. In this work, we revisit the overall model-based DRP objective and instead take a minorizationmaximization perspective to iteratively optimize the DRP w.r.t. a locally tight lower-bounded objective. This novel …

Go to article

Bug Triage Automation Approaches: A Comparative Study, Dr Khaled Nagaty, Madonna Mayez, Khaled Ahmed Nagay Dr. Mar 2022

Bug Triage Automation Approaches: A Comparative Study, Dr Khaled Nagaty, Madonna Mayez, Khaled Ahmed Nagay Dr.

Computer Science

Bug triage is an essential task in the software maintenance phase. It is the process of assigning a developer (fixer) to bug report. Triaging process is performed by the triager, who has to analyze developers’ profiles and bug reports for the purpose of making a suitable assignment. Manual assignment consumes time, financial resources and human resources; to get a high-quality software with minimum cost, automating this process is necessary. Previous researchers tackled this problem as a classification problem from different perspectives, either information retrieval approach or machine learning algorithms, some researchers handled it as an optimization problem using optimization and …

Go to article

A New Way To Trace Sars-Cov-2 Variants Through Weighted Network Analysis Of Frequency Trajectories Of Mutations, Paul W. Bible, Qiang Huang, Qiang Zhang, Qiaoxing Liang, Fangfang Zheng, Ying Wang, Yuantao Hao, Yu Liu Mar 2022

A New Way To Trace Sars-Cov-2 Variants Through Weighted Network Analysis Of Frequency Trajectories Of Mutations, Paul W. Bible, Qiang Huang, Qiang Zhang, Qiaoxing Liang, Fangfang Zheng, Ying Wang, Yuantao Hao, Yu Liu

Computer Science Faculty publications

Early detection of SARS-CoV-2 variants enables timely tracking of clinically important strains in order to inform the public health response. Current subtype-based variant surveillance depending on prior subtype assignment according to lag features and their continuous risk assessment may delay this process. We proposed a weighted network framework to model the frequency trajectories of mutations (FTMs) for SARS-CoV-2 variant tracing, without requiring prior subtype assignment. This framework modularizes the FTMs and conglomerates synchronous FTMs together to represent the variants. It also generates module clusters to unveil the epidemic stages and their contemporaneous variants. Eventually, the module-based variants are assessed by …

Go to article

Directional Pairwise Class Confusion Bias And Its Mitigation, Sudhashree Sayenju, Ramazan Aygun Phd, Jonathan Boardman, Duleep Prasanna Rathgamage Don, Yifan Zhang Phd, Bill Franks, Sereres Johnston Phd, George Lee, Dan Sullivan, Girish Modgil Phd Mar 2022

Directional Pairwise Class Confusion Bias And Its Mitigation, Sudhashree Sayenju, Ramazan Aygun Phd, Jonathan Boardman, Duleep Prasanna Rathgamage Don, Yifan Zhang Phd, Bill Franks, Sereres Johnston Phd, George Lee, Dan Sullivan, Girish Modgil Phd

Published and Grey Literature from PhD Candidates

Recent advances in Natural Language Processing have led to powerful and sophisticated models like BERT (Bidirectional Encoder Representations from Transformers) that have bias. These models are mostly trained on text corpora that deviate in important ways from the text encountered by a chatbot in a problem-specific context. While a lot of research in the past has focused on measuring and mitigating bias with respect to protected attributes (stereotyping like gender, race, ethnicity, etc.), there is lack of research in model bias with respect to classification labels. We investigate whether a classification model hugely favors one class with respect to another. …

Go to article

A Privacy-Preserving National Clinical Data Warehouse: Architecture And Analysis, Md Raihan Mia, Abu Sayed Md Latiful Hoque, Shahidul Islam Khan, Sheikh Iqbal Ahamed Mar 2022

A Privacy-Preserving National Clinical Data Warehouse: Architecture And Analysis, Md Raihan Mia, Abu Sayed Md Latiful Hoque, Shahidul Islam Khan, Sheikh Iqbal Ahamed

Computer Science Faculty Research and Publications

A centralized clinical data repository is essential for inspecting patients’ medical history, disease analysis, population-wide disease research, treatment decision support, and improving existing healthcare policies and services. Bangladesh, a rapidly developing country, poses several unusual challenges for developing such a centralized clinical data repository as the existing Electronic Health Records (EHR) are stored in unconnected, heterogeneous sources with no unique patient identifier and consistency. Data integration with secure record linkage, privacy preservation, quality control, and data standardization are the main challenges for developing a consistent and interoperable centralized clinical data repository. Based on the findings from our previous researches, we …

Go to article

First Prev Next Last