Wearable smart devices are capable of capturing a variety of information from their users using a multitude of noninvasive sensing modalities. Using features from the raw measurements of wearable devices, sensor fusio...
Wearable smart devices are capable of capturing a variety of information from their users using a multitude of noninvasive sensing modalities. Using features from the raw measurements of wearable devices, sensor fusion enables us to obtain a holistic picture of the users’ context and monitor their activity state with increased accuracy. Human activity recognition using noninvasive sensors allows us to capture the natural behavior of users in their day-to-day lives. This in-the-wild activity recognition, however, poses several key challenges that must be addressed to create effective classification models. The main challenges are class imbalance, uncertainty in classifier decisions, and large feature spaces. To address them, this study further explores a probabilistic sensor fusion method called Naive Adaptive Probabilistic Sensor (NAPS) Fusion. In doing so, we establish the viability of NAPS Fusion for natural human activity recognition using noninvasive sensing modalities. NAPS Fusion handles dimensionality reduction by creating reduced feature sets and mitigates the class imbalance issue through the use of Synthetic Minority Oversampling Technique (SMOTE). Moreover, NAPS Fusion addresses uncertainty in the decisions of classifiers using a Dempster-Shafer theoretic late fusion framework. Our empirical evaluation demonstrates that NAPS Fusion has broad applications beyond its original design for cognitive state detection. It outperforms similar decision level sensor fusion methods (late fusion using averaging, LFA, and late fusion using learned weights, LFL) in the detection of exercise and sedentary activities such as walking, running, lying down, and sitting. We observe improvements of up to 56% in F1 score and up to 59% in precision with NAPS Fusion over the compared methods.
Interest in operating commercial Urban Air Taxis (UAT) around the world has been growing rapidly over the last few years. One of the many challenges in designing aircraft suitable for operating in a turbulent urban ai...
详细信息
Interest in operating commercial Urban Air Taxis (UAT) around the world has been growing rapidly over the last few years. One of the many challenges in designing aircraft suitable for operating in a turbulent urban airflow environment is to design a robust inner loop flight controller. This study investigates the effect of filtered Angular Random Walk (ARW) error found in Inertial Measurement Units (IMU) on the inner loop flight controller's ability to maintain stable, wings level, horizontal flight, while not causing noticeable discomfort to passengers and respecting the limits of authority of the aircraft's control surfaces in a representative urban airflow environment. The performance of two controller architectures were investigated: classical Proportional, Integral, Derivative (PID) control scheme as well as Linear Active Disturbance Rejection control (LADRC) control scheme. The conclusion of this study provides recommendations on a minimum threshold of IMU sensor grades and general considersations that would be useful to the controller designer. The findings are demonstrated by observing the vertical acceleration, $n_{z}$ , angular rate setpoint tracking performance, and control surface deflections.
Projection-based controllers (PBCs) are currently gaining traction in both scientific and engineering communities. In PBCs, the input-output signals of the controller are kept in sector-bounded sets by means of projec...
Projection-based controllers (PBCs) are currently gaining traction in both scientific and engineering communities. In PBCs, the input-output signals of the controller are kept in sector-bounded sets by means of projection. In this paper, we will show how this projection operation can be used to induce useful passivity or general dissipativity properties for broad classes of (unprojected) nonlinear controllers that otherwise would not have these properties. The induced dissipativity properties of PBC will be exploited to guarantee asymptotic stability of negative feedback interconnections of passive nonlinear plants and suitably designed PBC, under mild conditions. Generalizations to so-called $(q,s,r)$ -dissipativity will be presented as well. To illustrate the effectiveness of PBC control design via these passivity-based techniques, a numerical example is provided.
In response to the pressing challenges in parking online reservation platforms, the primary issue this paper addresses is the need for a user-centric parking reservation experience. To tackle this problem, the study a...
In response to the pressing challenges in parking online reservation platforms, the primary issue this paper addresses is the need for a user-centric parking reservation experience. To tackle this problem, the study aims to develop a recommendation system that enhances user satisfaction and streamlines the parking reservation *** provide personalized parking recommendations, a hybrid multimodal recommendation system is designed, grounded in distance-based recommendation and content-based filtering, and taking into account user preferences and feedback, history behavior and proximity to preferred tourist attractions and points of *** leveraging a rich dataset comprising 1804 parking items, results indicate a notable improvement and more user-centric user experience, as the system suggests parking lots in line with user preferences and points of interest. User feedback mechanisms are seamlessly integrated, facilitating continuous adaptation and refinement based on user convenience and past *** work shows significant potential in enhancing user satisfaction and streamlining the user experience in parking online reservation systems.
Artificial intelligence (AI) has achieved great strides in recent years, with applications in a variety of areas of study, including healthcare. Consequently, the integration of artificial intelligence (AI) and medica...
Artificial intelligence (AI) has achieved great strides in recent years, with applications in a variety of areas of study, including healthcare. Consequently, the integration of artificial intelligence (AI) and medical imaging has ushered in a new era in healthcare diagnosis and therapy. Artificial intelligence (AI) has shown impressive potential in enhancing accuracy, efficiency, and diagnostic performance across a range of medical imaging modalities by using the power of deep learning (DL), machine learning (ML), and computer vision. In this paper, we are trying to investigate the connection between artificial intelligence (AI) and medical imaging, concentrating on how AI-driven strategies are improving performance at the cutting edge of medical imaging technologies through the proposed architecture model. Furthermore, the paper also explores the limitations and opportunities that result from incorporating artificial intelligence (AI) into the use of medical imaging. The potential for artificial intelligence (AI) to transform image-guided therapies and its implications for personalized medicine are investigated.
In new generation networks, 5G and 6G networks, intelligent mechanisms based on artificial intelligence algorithms are playing a relevant role in the performance improvement at different network levels. In 5G networks...
详细信息
Alzheimer's disease (AD) is a progressive brain disorder impacting behavior, memory, and cognition, with over a million cases reported annually in India. The risk significantly increases beyond age 65. Early diagn...
详细信息
ISBN:
(数字)9798331540821
ISBN:
(纸本)9798331540838
Alzheimer's disease (AD) is a progressive brain disorder impacting behavior, memory, and cognition, with over a million cases reported annually in India. The risk significantly increases beyond age 65. Early diagnosis and treatment can result in better recovery. We propose a predictive model using the Random Forest algorithm and the OASIS dataset for early AD diagnosis, leveraging MRI data, clinical notes, genetic markers, and cognitive test results. Our model was evaluated against several others, including Decision Tree, AdaBoost, SVM, and Logistic Regression. With a 97.3% accuracy and a 2.7% error rate, our Random Forest Classifier o utperformed t he others, demonstrating superior predictive power for early AD diagnosis and potentially improving patient care.
The development of cyber-attacks has unprecedented effect on businesses and governments. The recent years have witnessed various number of security breaches against organizations equipped by diverse cybersecurity solu...
详细信息
Internet of Things (IoT)-where the physical components are able to communicate with each other and with the internet-is one of the driving forces behind the Fourth Industrial Revolution. Nowadays, IoT applications hav...
详细信息
In an age of information and digital communication, social media platforms are becoming essential spaces where individuals can voice their thoughts, exchange ideas, and take part in discussions on a variety of subject...
详细信息
ISBN:
(数字)9798350379587
ISBN:
(纸本)9798350379594
In an age of information and digital communication, social media platforms are becoming essential spaces where individuals can voice their thoughts, exchange ideas, and take part in discussions on a variety of subjects. High quality and high value datasets of such ideas and discussions in textual and formatted form are fundamental to the research aspects of Artificial Intelligence and data-science. In this paper, we present a dataset of indirect harassment. There are approximately 10,700 tweets with binary labels. The labels were assigned by a team of researchers collectively. The dataset contains approximately 19 percent positive indirect harassment labels and 81 percent negative indirect harassment labels. The data is useful for training and running machine and deep learning models to detect indirect and direct harassment. The ease in understanding the data for researchers and other peers is also very crucial. The corpus mentioned in this paper is easy to understand via the use of a binary label. The data is simplistic in nature because there are only two columns with one being the text and the other being the indirect label. This work was necessary because it was a component of a larger project that also needed this kind of dataset.
暂无评论