检索结果-内蒙古大学图书馆

intelligent Near-Infrared Spectroscopy for Blood Glucose Level Classification

学校读者我要写书评

暂无评论

Intelligent Near-Infrared Spectroscopy for Blood Glucose Lev...

Iranian Conference on Electrical Engineering (ICEE)

作者： Shahrooz Sharifi Amirhoseein Maddah-T. Mohammad-R. Akbarzadeh-T. Department of Electrical Engineering Ferdowsi University of Mashhad Mashhad Iran Centre of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran

ISBN: (数字)9798350376340

ISBN: (纸本)9798350376357

Diabetes Mellitus is a disease characterized by inadequate control of blood glucose levels, and it ranks among the leading causes of human mortality, the measurement of which, requires fingertip pricking. Nowadays, non-invasive healthcare monitoring systems based on wearable sensors and machine learning models hold the future of smart health. However, the nature of these methods is such that they get affected by internal physiological and external parameters, which is one of the obstacles to this path. This study aims to introduce a non-invasive blood glucose level classification method based on machine learning analysis. The device's sensor consists of an optical finger sensor that transmits near-infrared light, filtering, and amplification to reduce the noise of the extracted photoplethysmography signal. Moreover, we have adopted the four-stage framework of biomedical signal processing to analyze the acquired PPG signal. Before using the Savitzky-Golay derivation to pre-filter the signal and prepare it for feature extraction, it was normalized using Standard Normal Variate (SNV). In addition, four different machine learning models, including Support Vector Machine (SVM), Weighted K-Nearest Neighbors (KNN), Wide Neural Network, and Decision Tree were used for blood glucose level classification. For this study, a dataset was created consisting of 106 data, gathered from 27 subjects. Findings revealed that the Weighted KNN exhibited the best performance among other classification models, having 90.5% accuracy.

关键词： Support vector machines Stimulated emission Smart healthcare Machine learning Nearest neighbor methods Feature extraction Glucose

Robust Non-parametric Knowledge-based Diffusion Least Mean Squares over Adaptive Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ashkezari-Toussi, Soheila Sadoghi-Yazdi, Hadi Department of Computer Engineering Ferdowsi University of Mashhad Mashhad Iran Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran

The present study proposes incorporating of non-parametric knowledge to the diffusion least-mean-squares algorithm in the framework of a maximum a posteriori (MAP) estimation. The proposed algorithm leads to a robust estimation of an unknown parameter vector in a group of cooperative estimators. Utilizing kernel density estimation and buffering some intermediate estimations, the prior distribution and conditional likelihood of the parameters vector in each node are calculated. Pseudo Huber loss function is used for designing the likelihood function. Also, an error thresholding function is defined to reduce the computational overhead as well as more relaxation against noise, which stops the update every time an error is less than a predefined threshold. The performance of the proposed algorithm is examined in the stationary and non-stationary scenarios in the presence of Gaussian and non-Gaussian noise. Results show the robustness of the proposed algorithm in the presence of different noise types. Copyright © 2023, The Authors. All rights reserved.

关键词： Gaussian noise (electronic)

Robust Human Movement Prediction by Completion-Generative Adversarial Networks with Huber Loss

学校读者我要写书评

暂无评论

Robust Human Movement Prediction by Completion-Generative Ad...

Iranian Conference of Biomedical Engineering (ICBME)

作者： Mojgan Azari Hamed Rafiei Mohammad-R Akbarzadeh-T Department of Electrical Engineering Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran

In recent years, wearable exoskeleton robots have been growingly used for rehabilitation or movement assistive purposes. Despite the growing application of these robots in various domains, such as physical therapy, the movement synchronization between robots and human bodies remains a challenging problem. This paper aims to achieve better synchronization by predicting human movement. Although several works have been presented in this domain, the robustness of these predictions has received less attention. This paper aims to provide a robust prediction using Completion-Generative Adversarial Networks (CGAN) that are learned based on the Huber loss function. Specifically, we reshape the 3D-joint-position-time series (jointxaxesxtime) into multivariate time series ((jointxaxes) xtime) and pass them to a CGAN. We use the Huber loss function to improve the GAN performance and offer higher robustness against noise in real-world applications. The proposed method is evaluated on an actual human gait dataset and compared with several recent works in this domain. Results show that the proposed method is superior to the previous works in prediction error, particularly in terms of achieving a better signal-to-noise ratio.

关键词： Time series analysis Robot control Exoskeletons Medical treatment Generative adversarial networks Robustness Generators

Shadfa 0.1: The Iranian Movie Knowledge Graph and Graph-Embedding-Based Recommender System

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Pouyan, Rayhane Kalamati, Hadi Ebrahimian, Hannane Karrabi, Mohammad Akbarzadeh, Mohammad R.T. Department of Computer engineering Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran School of Electrical and Computer Engineering Faculty of Engineering University of Tehran Tehran Iran

Movies are a great source of entertainment. However, the problem arises when one is trying to find the desired content within this vast amount of data which is significantly increasing every year. Recommender systems (RSs) can provide appropriate algorithms to solve this problem. The content-based technique has found popularity due to the lack of available user data in most cases. Content-based RSs are based on similarity of items demographic information;Term Frequency – Inverse Document Frequency (TF-IDF) and Knowledge Graph Embedding (KGE) are two approaches used to vectorize data to calculate these similarities. In this paper, we propose a weighted content-based movie RS by combining TF-IDF which is an appropriate approach for embedding textual data such as plot/description and KGE which is used to embed named entities such as director’s name. The weights between features are determined using a Genetic algorithm. Additionally, Iranian movies dataset is created by scraping data from movie-related websites. This dataset and the structure of the FarsBase KG are used to create the MovieFarsBase KG which is a component in the implementation process of the proposed content-based RS. Using precision, recall, and F1 score metrics, this study shows that the proposed approach outperforms the conventional approach that uses TF-IDF for embedding all attributes. © 2022, CC BY.

关键词： Genetic algorithms

Cooperative Fuzzy Adaptive Control for Stochastic Nonlinear Multi-Agent Systems Using Artificial Potential Functions

学校读者我要写书评

暂无评论

Cooperative Fuzzy Adaptive Control for Stochastic Nonlinear ...

IEEE International Conference on Systems, Man and Cybernetics

作者： F. Baghbani P. Parsa M.-R. Akbarzadeh-T Semnan University Semnan Iran Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran

ISBN: (纸本)9781665442084

A realistic control strategy for stochastic multi-agent systems (MAS) should consider uncertainties such as unknown dynamics of the agents, limited accessibility of the leader state, and external disturbances. Here, we propose an adaptive fuzzy control structure for a class of stochastic uncertain MAS that have partially unknown heterogeneous higher-order dynamics of the agents. Artificial potential functions (APFs) are incorporated to model the communication among the agents, and the unknown dynamics are approximated by the aid of fuzzy logic systems. The current work extends our previous work to the case where only some of the agents could access the states of the leader to reach an n-dimensional consensus on the leader states. The overall system’s stability is assured using Lyapunov stability theory, and all of the closed-loop signals of the system are shown to be semi-globally uniformly ultimately bounded (SGUUB) in the mean square. Simulation results ensure the capability of the system in low tracking error with reasonable control input usage.

关键词： Fuzzy logic Fuzzy control Uncertainty Simulation Stochastic processes Stability analysis Trajectory

Solving a Multi-Traveling Salesmen Problem using a Mamdani Fuzzy Inference Engine and Simulated Annealing Search Algorithm 10

学校读者我要写书评

暂无评论

Solving a Multi-Traveling Salesmen Problem using a Mamdani F...

10th International Conference on Computer and Knowledge Engineering, ICCKE 2020

作者： Hassanpour, Fatemeh Akbarzadeh-T, Mohamad-R. Ferdowsi University Scientific Center of Soft Computing and Intelligent Information Processing Department of Computer Engineering Mashhad Iran

ISBN: (纸本)9781728185668

The multi-traveling salesmen problem (MTSP) is an extended situation of the standard traveling salesman problem (TSP), in which there is more than one salesman. In this matter, several salesmen are determined to visit N city with the goal of the shortest route to selling their goods, assuming they have crossed all of them, via just once each. The aim is to minimize the total cost of travel for salesman. Thus, it can be modeled as an optimization problem. Regarding the complexity degree, this problem is well known as a NP-Hard problem. Therefore, several meta-heuristic algorithms have been developed at the frontiers of knowledge to solve this problem. Nevertheless, the computational and time complexities are the most important challenges of such algorithms. In this paper, we first convert the MTSP using a fuzzy approach with a linear complexity level, to several TSPs. Then, we solve each problem using the simulated annealing (SA) optimization algorithm. In this way, the time complexity of the system is significantly reduced using the proposed method as well as the accuracy of the system is satisfactory. To assess the proposed algorithm, this method is implemented in the TSPLIB library dataset. © 2020 IEEE.

关键词： Simulated annealing

Kineticomyography for Hand Gesture Recognition: Experimental Implementation

学校读者我要写书评

暂无评论

Kineticomyography for Hand Gesture Recognition: Experimental...

Iranian Conference on intelligent Systems (ICIS)

作者： Babak Sarani Mohammad-R Gholami-Z Hojjat Hosseini Mohammad-R. Akbarzadeh-T A. Akbarzadeh Dept. of Mechanical Engineering Center of Advanced Rehabilitation and Robotics Research(FUM CARE) Ferdowsi University of Mashhad(FUM) Mashhad Iran Dept. of Electrical Engineering Center of Advanced Rehabilitation and Robotics Research(FUM CARE) Ferdowsi University of Mashhad(FUM) Mashhad Iran Dept. of Electrical Engineering Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad(FUM) Mashhad Iran

ISBN: (数字)9798331507565

ISBN: (纸本)9798331507572

Human-machine interfaces (HMIs) play a vital role in medical science and technology by interpreting human voluntary movements, with ElectroMyoGraphy (EMG) being the most widely used signal. This paper explores KineticoMyoGra- phy (KMG) as an innovative alternative, offering greater noise robustness, simpler connectivity, and more intuitive movement control. We evaluated the online implementation of KMG signals for bionic hand control using a 3-axis simulator that mimics muscle movements through three magnets in two different data acquisition scenarios. The experiment investigated five distinct classes of forearm muscle movements, including a rest class, and KMG signals were recorded using 40 sensors across four custom-made PCB boards. After removing signal spikes, various models were trained, and the top five were selected for further evaluation. In online testing, the Support Vector Machine (SVM) achieved 90% accuracy in the first scenario, while the Deep Neural Network (DNN) reached 77% accuracy in the second scenario.

关键词： Hands Support vector machines Accuracy Neuromuscular Artificial neural networks Electromyography Recording Noise robustness Magnets Testing

Adaptive Locally Linear Embedding

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Goli, Ali Alizadeh, Mahdieh Yazdi, Hadi Sadoghi Department of Computer Engineering Ferdowsi University of Mashhad Mashhad Iran Center of Excellence in Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran

Manifold learning techniques, such as Locally linear embedding (LLE), are designed to preserve the local neighborhood structures of high-dimensional data during dimensionality reduction. Traditional LLE employs Euclidean distance to define neighborhoods, which can struggle to capture the intrinsic geometric relationships within complex data. A novel approach, Adaptive locally linear embedding(ALLE), is introduced to address this limitation by incorporating a dynamic, data-driven metric that enhances topological preservation. This method redefines the concept of proximity by focusing on topological neighborhood inclusion rather than fixed distances. By adapting the metric based on the local structure of the data, it achieves superior neighborhood preservation, particularly for datasets with complex geometries and high-dimensional structures. Experimental results demonstrate that ALLE significantly improves the alignment between neighborhoods in the input and feature spaces, resulting in more accurate and topologically faithful embeddings. This approach advances manifold learning by tailoring distance metrics to the underlying data, providing a robust solution for capturing intricate relationships in high-dimensional datasets. © 2025, CC BY-SA.

关键词： Topology

A Quantum-like Model for Predicting Human Decisions in the Entangled Social Systems

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Meghdadi, Aghdas Akbarzadeh, Mohammad R.T. Javidan, Kurosh The Department of Electrical Engineering Center of Excellence on Soft Computing and Intelligent Information Processing Ferdowsi University of Mashhad Mashhad Iran The Department of Physics Ferdowsi University of Mashhad Mashhad Iran

Human-centered systems of systems such as social networks, the Internet of Things, or healthcare systems are growingly becoming significant facets of modern life. Realistic models of human behavior in such systems play an essential role in their accurate modeling and prediction. Nevertheless, human behavior under uncertainty often violates the predictions by the conventional probabilistic models. Recently, quantum-like decision theories have shown a considerable potential to explain the contradictions in human behavior by applying quantum probabilities. But providing a quantum-like decision theory that could predict rather than describe the current state of human behavior is still one of the unsolved challenges. The fundamental contribution of this work is introducing the concept of entanglement from quantum information theory to Bayesian networks. This concept leads to an entangled quantum-like Bayesian network, in which each human is a part of the entire society. Accordingly, society's effect on the dynamic evolution of the decision-making process, which is less often considered in decision theories, is modeled by entanglement measures. To reach this aim, we introduce a quantum-like witness and find the relationship between this witness and the famous concurrence entanglement measure. The proposed predictive entangled quantum-like Bayesian network (PEQBN) is evaluated on 22 experimental tasks. Results confirm that PEQBN provides more realistic predictions of human decisions under uncertainty when compared with classical Bayesian networks and three recent quantum-like approaches. © 2021, CC BY-NC-ND.

关键词： Decision making