-
Understanding the Relationship Between Personal Data Privacy Literacy and Data Privacy Information Sharing by University Students
Authors:
Brady D. Lund,
Bryan Anderson,
Ana Roeschley,
Gahangir Hossain
Abstract:
With constant threats to the safety of personal data in the United States, privacy literacy has become an increasingly important competency among university students, one that ties intimately to the information sharing behavior of these students. This survey based study examines how university students in the United States perceive personal data privacy and how their privacy literacy influences th…
▽ More
With constant threats to the safety of personal data in the United States, privacy literacy has become an increasingly important competency among university students, one that ties intimately to the information sharing behavior of these students. This survey based study examines how university students in the United States perceive personal data privacy and how their privacy literacy influences their understanding and behaviors. Students responses to a privacy literacy scale were categorized into high and low privacy literacy groups, revealing that high literacy individuals demonstrate a broader range of privacy practices, including multi factor authentication, VPN usage, and phishing awareness, whereas low literacy individuals rely on more basic security measures. Statistical analyses suggest that high literacy respondents display greater diversity in recommendations and engagement in privacy discussions. These findings suggest the need for enhanced educational initiatives to improve data privacy awareness at the university level to create a better cyber safe population.
△ Less
Submitted 24 May, 2025;
originally announced May 2025.
-
Learning Patterns in Imaginary Vowels for an Intelligent Brain Computer Interface (BCI) Design
Authors:
Parisa Ghane,
Gahangir Hossain
Abstract:
Technology advancements made it easy to measure non-invasive and high-quality electroencephalograph (EEG) signals from human's brain. Hence, development of robust and high-performance AI algorithms becomes crucial to properly process the EEG signals and recognize the patterns, which lead to an appropriate control signal. Despite the advancements in processing the motor imagery EEG signals, the hea…
▽ More
Technology advancements made it easy to measure non-invasive and high-quality electroencephalograph (EEG) signals from human's brain. Hence, development of robust and high-performance AI algorithms becomes crucial to properly process the EEG signals and recognize the patterns, which lead to an appropriate control signal. Despite the advancements in processing the motor imagery EEG signals, the healthcare applications, such as emotion detection, are still in the early stages of AI design. In this paper, we propose a modular framework for the recognition of vowels as the AI part of a brain computer interface system. We carefully designed the modules to discriminate the English vowels given the raw EEG signals, and meanwhile avoid the typical issued with the data-poor environments like most of the healthcare applications. The proposed framework consists of appropriate signal segmentation, filtering, extraction of spectral features, reducing the dimensions by means of principle component analysis, and finally a multi-class classification by decision-tree-based support vector machine (DT-SVM). The performance of our framework was evaluated by a combination of test-set and resubstitution (also known as apparent) error rates. We provide the algorithms of the proposed framework to make it easy for future researchers and developers who want to follow the same workflow.
△ Less
Submitted 18 February, 2022; v1 submitted 17 October, 2020;
originally announced October 2020.
-
A Fourier Domain Feature Approach for Human Activity Recognition & Fall Detection
Authors:
Asma Khatun,
Sk. Golam Sarowar Hossain
Abstract:
Elder people consequence a variety of problems while living Activities of Daily Living (ADL) for the reason of age, sense, loneliness and cognitive changes. These cause the risk to ADL which leads to several falls. Getting real life fall data is a difficult process and are not available whereas simulated falls become ubiquitous to evaluate the proposed methodologies. From the literature review, it…
▽ More
Elder people consequence a variety of problems while living Activities of Daily Living (ADL) for the reason of age, sense, loneliness and cognitive changes. These cause the risk to ADL which leads to several falls. Getting real life fall data is a difficult process and are not available whereas simulated falls become ubiquitous to evaluate the proposed methodologies. From the literature review, it is investigated that most of the researchers used raw and energy features (time domain features) of the signal data as those are most discriminating. However, in real life situations fall signal may be noisy than the current simulated data. Hence the result using raw feature may dramatically changes when using in a real life scenario. This research is using frequency domain Fourier coefficient features to differentiate various human activities of daily life. The feature vector constructed using those Fast Fourier Transform are robust to noise and rotation invariant. Two different supervised classifiers kNN and SVM are used for evaluating the method. Two standard publicly available datasets are used for benchmark analysis. In this research, more discriminating results are obtained applying kNN classifier than the SVM classifier. Various standard measure including Standard Accuracy (SA), Macro Average Accuracy (MAA), Sensitivity (SE) and Specificity (SP) has been accounted. In all cases, the proposed method outperforms energy features whereas competitive results are shown with raw features. It is also noticed that the proposed method performs better than the recently risen deep learning approach in which data augmentation method were not used.
△ Less
Submitted 11 March, 2020;
originally announced March 2020.
-
Open Challenges and Issues: Artificial Intelligence for Transactive Management
Authors:
Asma Khatun,
Sk. Golam Sarowar Hossain
Abstract:
The advancement of Artificial Intelligence (AI) has improved the automation of energy managements. In smart energy management or in a smart grid framework, all the devices and the distributed resources and renewable resources are embedded which leads to reduce cost. A smart energy management system, Transactive management (TM) is a concept to improve the efficiency and reliability of the power sys…
▽ More
The advancement of Artificial Intelligence (AI) has improved the automation of energy managements. In smart energy management or in a smart grid framework, all the devices and the distributed resources and renewable resources are embedded which leads to reduce cost. A smart energy management system, Transactive management (TM) is a concept to improve the efficiency and reliability of the power system. The aim of this article is to look for the current development of TM methods based on AI and Machine Learning (ML) technology. In AI paradigm, MultiAgent System (MAS) based method is an active research area and are still in evolution. Hence this article describes how MAS based method applied in TM. This paper also finds that MAS based method faces major difficulty to design or set up goal to various agents and describes how ML technique can contribute to that solution. A brief comparison analysis between MAS and ML techniques are also presented. At the end, this article summarizes the most relevant open challenges and issues on the AI based methods for transactive energy management.
△ Less
Submitted 2 January, 2020;
originally announced January 2020.
-
Early Detection of Diabetic Retinopathy and Severity Scale Measurement: A Progressive Review & Scopes
Authors:
Asma Khatun,
Sk. Golam Sarowar Hossain
Abstract:
Early detection of diabetic retinopathy prevents visual loss and blindness of a human eye. Based on the types of feature extraction method used, DR detection method can be broadly classified as Deep Convolutional Neural Network (CNN) based and traditional feature extraction (machine learning) based. This paper presents a comprehensive survey of existing feature extraction methods based on Deep CNN…
▽ More
Early detection of diabetic retinopathy prevents visual loss and blindness of a human eye. Based on the types of feature extraction method used, DR detection method can be broadly classified as Deep Convolutional Neural Network (CNN) based and traditional feature extraction (machine learning) based. This paper presents a comprehensive survey of existing feature extraction methods based on Deep CNN and conventional feature extraction for DR detection. In addition to that, this paper focuses on the severity scale measurement of the DR detection and to the best of our knowledge this is the first survey paper which covers severity grading scale. It is also necessary to mention that this is the first study which reviews the proposed Deep CNN based method in the state of the art for DR detection methods. This study discovers that recently proposed deep learning based DR detection methods provides higher accuracy than existing traditional feature extraction methods in the literature and also useful in large scale datasets. However, deep learning based methods require GPU implementation to get the desirable output. The one of the other major finding of this paper is that there are no obvious standard severity scale detection criteria to measure the grading. Some used binary class while many other used multi stage class.
△ Less
Submitted 30 December, 2019;
originally announced December 2019.