Withtheir ability to analyse and learn from complex data, neural networks have emerged as an ubiquitous machine learning tool with applications in many industries. Despite the success of neural networks in various ta...
Withtheir ability to analyse and learn from complex data, neural networks have emerged as an ubiquitous machine learning tool with applications in many industries. Despite the success of neural networks in various tasks, obtaining the optimal structure for a specific problem is both time-consuming and difficult because of the large number of hyperparameters that need to be properly selected. To address this issue, this work proposes an approach for automating the design of neural networks through the use of genetic algorithms (GAs). More specifically, the proposed GA-based technique aims to find the optimal Convolutional Neural Network (CNN) architecture and learning-related hyperparameters for the problem of image classification. the approach is evaluated on the MNIST dataset, a widely used benchmark for image classification tasks, on which our top performing architecture obtains an accuracy of 99.35%. these results demonstrate the effectiveness of the proposed approach in terms of finding architectures that outperform those designed by human experts and are competitive with state-of-the-art models. the evaluation results are accompanied by a comprehensive analysis of the evolution of hyperparameters within generations and the changes in network architecture over time, providing insights into how the GA explores the search space. the proposed approach is a step forward in the direction of automating the process of neural network design and its results indicate that it can be a valuable tool for researchers.
A Digital Twin is envisioned as a natural evolution with regard to the integration of physical and virtual environments with application in various domains such as organization management, products, services, e-health...
A Digital Twin is envisioned as a natural evolution with regard to the integration of physical and virtual environments with application in various domains such as organization management, products, services, e-health, business intelligence. It consists of a virtual model of a physical object, which dynamically pairs the physical entity with it’s digital replica. the e-health applications are heterogeneous and complex by nature, having a great variability, so is difficult to be managed and monitored through a unique ICT system. A Digital Twin is envisioned to connect the physical world associated to healthcare and the cyber world. Relevant patient and hospital environment information are stored in the Electronic Health Record (EHR). there is an expectation that the traditional Electronic Medical Records, EHR, will be enhanced or even replaced by Digital Twins. this paper discusses a generic Digital Twin based approach in healthcare, which extends the EHR capabilities, within a framework of interoperability that enables the possibility of developing complex e-health applications is presented. Physical collected healthcare information, EHR data and Digital Twin processed outputs are available as services for the users, within a Service Oriented Architecture. the data traffic withthe Digital Twin – EHR framework is realized through Web services, which are able to assure security, safety and privacy of the medical transmitted information, in a QoS approach.
Within a partnership between the Carol Davila University of Medicine and Pharmacy in Bucharest and the University Politehnica of Bucharest, an educational web-based platform is being developed. the platform enables me...
Within a partnership between the Carol Davila University of Medicine and Pharmacy in Bucharest and the University Politehnica of Bucharest, an educational web-based platform is being developed. the platform enables medical students to virtually reproduce the interactions that they usually have with patients in hospitals, through an avatar in clinically valid scenarios. Students can interact with virtual patients through text and speech interactions. they can access the health records of the virtual patients, perform objective exams, request and analyze medical investigations, elaborate treatment plans and provide diagnoses. they can receive feedback from a professor in real time or after the end of the session. the platform integrates various tools that professors can use to provide feedback, generate scenarios, manage the medical data on the platform and annotate the data. In the current stage of the implementation, the platform supports English and Romanian languages, it incorporates some customizable features, a user ranking system and a scoring system that evaluates and scores each session of the student and rewards the student with points. the scoring system takes into consideration the feedback provided by the professor, who may manually adjust the final score (if he/she desires).
the Internet of things(IoT)has become an expansion of connected smart devices from small sensors to complex fog and cloud nodes and various network technologies and communication protocols. Devices, including Intellig...
详细信息
Considering the constant growth and evolution of technology, being able to secure data is a challenge. this is why researchers find this a top research topic. Cyber-Physical systems are complex, large-scale systemsth...
Considering the constant growth and evolution of technology, being able to secure data is a challenge. this is why researchers find this a top research topic. Cyber-Physical systems are complex, large-scale systemsthat can be found in domains like healthcare, transportation, industrial controls systems or smart grids. Having many components that interact with each other, we can observe here the challenge in maintaining a secure environment for the data as data tampering or data extraction can occur, for example, in the actual transportation of data from one device to another. the main focus of this paper is to give an answer if blockchain technology provides a solution in building a secure network. We give a short review of the related work regarding blockchain integration in large-scale systems like Cyber-Physical systems. It was also briefly analyzed the layered structure of a blockchain and discussed some solutions for each layer. We then discussed about consensus algorithms for validation and authentication process. In general, consensus algorithms are implemented to assure the validation process of the nodes in a network. In this paper we tried to analyze and propose a model that offers an extra layer of security by adding the authentication step. In this way, the nodes that are validated are also considered trusted nodes in the network.
the proceedings contain 16 papers. the special focus in this conference is on Product-Focused Software Process Improvement. the topics include: Speech therapy Supported by AI and Smart Assistants;integration of Classi...
ISBN:
(纸本)9783031492655
the proceedings contain 16 papers. the special focus in this conference is on Product-Focused Software Process Improvement. the topics include: Speech therapy Supported by AI and Smart Assistants;integration of Classical and Quantum Services Using an Enterprise Service Bus;problem Decomposition to Leverage Quantum Computing for Optimization Problems;quantum Algorithm Cards: Streamlining the Development of Hybrid Classical-Quantum Applications;simulation-Based Safety Testing of Automated Driving systems;utilization of Machine Learning for the Detection of Self-admitted Vulnerabilities;closing the Loop: Towards a Complete Metamorphic Testing Pipeline;declarative Process Mining for Software Processes: the RuM Toolkit and the Declare4Py Python Library;MaREA: Multi-class Random Forest for Automotive Intrusion Detection;forecasting the Developer’s Impact in Managing the Technical Debt;development of a Desktop Application to Enable Doctors to Remotely Monitor Patients’ Hematological Parameters;the Significance of Classical Simulations in the Adoption of Quantum Technologies for Software Development;Enhancing Bug-Fixing Time Prediction with LSTM-Based Approach;enhancing Code Obfuscation Techniques: Exploring the Impact of Artificial Intelligence on Malware Detection.
Voltage control becomes a crucial challenge as the penetration of renewable energy sources (RESs) increases in distribution systems. the intermittent and variable nature of RESs like solar and wind power can lead to s...
详细信息
Skin cancer is one of the most common types of cancer, and it is caused by a variety of dermatological conditions. Identifying abnormalities from skin images is an important pre-diagnostic step to assist physicians in...
Skin cancer is one of the most common types of cancer, and it is caused by a variety of dermatological conditions. Identifying abnormalities from skin images is an important pre-diagnostic step to assist physicians in determining the patient’s condition. thus, to aid dermatologists in the diagnosis process, we proposed five CNN-based classification approaches namely ResNet-101, DenseNet-121, GoogLeNet, VGG16, and MobileNetV2 architectures on which the transfer learning process was applied. the HAM10000-N database consisting of 7,120 images, which was obtained from the original HAM10000 dataset through an augmentation process, was used to train the proposed methods. Moreover, the images from the HAM10000-N were pre-processed by removing hair withthe DullRazor algorithm. To evaluate and compare the performance of all networks five metrics were calculated: accuracy, precision, recall, and Fl-score. the best results for the seven-class classification of the HAM10000-N dataset were obtained for DenseNet-121 architecture with 87% accuracy, 0.871 precision, 0.87 recall and 0.872 F1-score.
Connected and automated vehicles (CAVs) have great potential to smooth mixed traffic flow. To focus on the fact that the driving styles of HDVs are time-varying, a new control framework based on time-varying model pre...
详细信息
the proceedings contain 16 papers. the special focus in this conference is on Product-Focused Software Process Improvement. the topics include: Speech therapy Supported by AI and Smart Assistants;integration of Classi...
ISBN:
(纸本)9783031492686
the proceedings contain 16 papers. the special focus in this conference is on Product-Focused Software Process Improvement. the topics include: Speech therapy Supported by AI and Smart Assistants;integration of Classical and Quantum Services Using an Enterprise Service Bus;problem Decomposition to Leverage Quantum Computing for Optimization Problems;quantum Algorithm Cards: Streamlining the Development of Hybrid Classical-Quantum Applications;simulation-Based Safety Testing of Automated Driving systems;utilization of Machine Learning for the Detection of Self-admitted Vulnerabilities;closing the Loop: Towards a Complete Metamorphic Testing Pipeline;declarative Process Mining for Software Processes: the RuM Toolkit and the Declare4Py Python Library;MaREA: Multi-class Random Forest for Automotive Intrusion Detection;forecasting the Developer’s Impact in Managing the Technical Debt;development of a Desktop Application to Enable Doctors to Remotely Monitor Patients’ Hematological Parameters;the Significance of Classical Simulations in the Adoption of Quantum Technologies for Software Development;Enhancing Bug-Fixing Time Prediction with LSTM-Based Approach;enhancing Code Obfuscation Techniques: Exploring the Impact of Artificial Intelligence on Malware Detection.
暂无评论