Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

Wyszukujesz frazę "learning network" wg kryterium: Temat


Tytuł:
An Analysis of Novel Money Laundering Data Using Heterogeneous Graph Isomorphism Networks. FinCEN Files Case Study
Wykorzystanie heterogenicznych grafowych sieci izomorficznych w analizie danych związanych z praniem brudnych pieniędzy. Studium przypadku FinCEN
Autorzy:
Wójcik, Filip
Powiązania:
https://bibliotekanauki.pl/articles/38890419.pdf
Data publikacji:
2024
Wydawca:
Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu
Tematy:
money laundering
deep learning
machine learning
network analysis
graphs
pranie brudnych pieniędzy
uczenie głębokie
analiza sieci
grafy
Opis:
Aim: This study aimed to develop and apply the novel HexGIN (Heterogeneous extension for Graph Isomorphism Network) model to the FinCEN Files case data and compare its performance with existing solutions, such as the SAGE-based graph neural network and Multi-Layer Perceptron (MLP), to demonstrate its potential advantages in the field of anti-money laundering systems (AML). Methodology: The research employed the FinCEN Files case data to develop and apply the HexGIN model in a beneficiary prediction task for a suspicious transactions graph. The model's performance was compared with the existing solutions in a series of cross-validation experiments. Results: The experimental results on the cross-validation data and test dataset indicate the potential advantages of HexGIN over the existing solutions, such as MLP and Graph SAGE. The proposed model outperformed other algorithms in terms of F1 score, precision, and ROC AUC in both training and testing phases. Implications and recommendations: The findings demonstrate the potential of heterogeneous graph neural networks and their highly expressive architectures, such as GIN, in AML. Further research is needed, in particular to combine the proposed model with other existing algorithms and test the solution on various money-laundering datasets. Originality/value: Unlike many AML studies that rely on synthetic or undisclosed data sources, this research was based on a publicly available, real, heterogeneous transaction dataset, being part of a larger investigation. The results indicate a promising direction for the development of modern hybrid AML tools for analysing suspicious transactions; based on heterogeneous graph networks capable of handling various types of entities and their connections.
Cel: Celem niniejszej analizy jest opracowanie i zastosowanie nowego modelu HexGIN (heterogeniczne rozszerzenie dla izomorfizmu sieci grafowych) do danych z dochodzenia dziennikarskiego FinCEN oraz porównanie jego jakości predykcji z istniejącymi rozwiązaniami, takimi jak sieć SAGE i wielowarstwowa sieć neuronowa (MLP). Metodyka: W badaniach wykorzystano dane ze śledztwa FinCEN do opracowania i zastosowania modelu HexGIN w zadaniu przewidywania beneficjenta sieci powiązanych transakcji finansowych. Skuteczność modelu porównano z istniejącymi rozwiązaniami wykorzystującymi sieci neuronowe grafu w serii eksperymentów z walidacją krzyżową. Wyniki: Eksperymentalne wyniki na danych walidacji krzyżowej i zestawie testowym potwierdzają potencjalne zalety HexGIN w porównaniu z istniejącymi rozwiązaniami, takimi jak MLP i SAGE. Proponowany model przewyższa inne algorytmy pod względem wyniku miary F1, precyzji i ROC AUC, w fazie zarówno treningowej, jak i testowej. Implikacje i rekomendacje: Wyniki pokazują potencjał heterogenicznych grafowych sieci i ich wysoce ekspresyjnych implementacji, takich jak GIN, w analizie transakcji finansowych. Potrzebne są dalsze badania, zwłaszcza w celu połączenia proponowanego modelu z innymi istniejącymi algorytmami i przetestowania rozwiązania na różnych zestawach danych dotyczących problemu prania brudnych pieniędzy. Oryginalność/wartość: W przeciwieństwie do wielu badań, które opierają się na syntetycznych lub nieujawnionych źródłach danych związanych z praniem brudnych pieniędzy, to studium przypadku opiera się na publicznie dostępnych, rzeczywistych, heterogenicznych danych transakcyjnych, będących częścią większego śledztwa dziennikarskiego. Wyniki wskazują obiecujący kierunek dla rozwoju nowoczesnych hybrydowych narzędzi do analizy podejrzanych transakcji, opartych na heterogenicznych sieciach grafowych.
Źródło:
Econometrics. Ekonometria. Advances in Applied Data Analytics; 2024, 28, 2; 32-49
1507-3866
Pojawia się w:
Econometrics. Ekonometria. Advances in Applied Data Analytics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Artificial intelligence-powered pulse sequences in nuclear magnetic resonance and magnetic resonance imaging: historical trends, current innovations and perspectives
Autorzy:
Tokarz, Paweł
Powiązania:
https://bibliotekanauki.pl/articles/35508129.pdf
Data publikacji:
2024
Wydawca:
Radomskie Towarzystwo Naukowe
Tematy:
artificial intelligence
machine learning
evolutionary algorithm
artificial neural network
nuclear magnetic resonance
magnetic resonance imaging
pulse sequence
shaped pulse
sztuczna inteligencja
uczenie maszynowe
algorytm ewolucyjny
sztuczna sieć neuronowa
magnetyczny rezonans jądrowy
rezonans magnetyczny
sekwencja impulsów
impuls ukształtowany
Opis:
This review article explores the historical background and recent advances in the application of artificial intelligence (AI) in the development of radiofrequency pulses and pulse sequences in nuclear magnetic resonance spectroscopy (NMR) and imaging (MRI). The introduction of AI into this field, which traces back to the late 1970s, has recently witnessed remarkable progress, leading to the design of specialized frameworks and software solutions such as DeepRF, MRzero, and GENETICS-AI. Through an analysis of literature and case studies, this review tracks the transformation of AI-driven pulse design from initial proof-of-concept studies to comprehensive scientific programs, shedding light on the potential implications for the broader NMR and MRI communities. The fusion of artificial intelligence and magnetic resonance pulse design stands as a promising frontier in spectroscopy and imaging, offering innovative enhancements in data acquisition, analysis, and interpretation across diverse scientific domains.
Źródło:
Scientiae Radices; 2024, 3, 1; 30-52
2956-4808
Pojawia się w:
Scientiae Radices
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Convolutional neural networks in the SSI analysis for mine-induced vibrations
Autorzy:
Zając, Maciej
Kuźniar, Krystyna
Powiązania:
https://bibliotekanauki.pl/articles/38707462.pdf
Data publikacji:
2024
Wydawca:
Instytut Podstawowych Problemów Techniki PAN
Tematy:
deep learning
convolutional neural network
shallow neural network
small data set
soil-structure interaction
mine-induced vibrations
głęboka nauka
splotowa sieć neuronowa
płytka sieć neuronowa
mały zestaw danych
interakcja gleba-struktura
wibracje wywołane minami
Opis:
Deep neural networks (DNNs) have recently become one of the most often used softcomputational tools for numerical analysis. The huge success of DNNs in the field of imageprocessing is associated with the use of convolutional neural networks (CNNs). CNNs,thanks to their characteristic structure, allow for the effective extraction of multi-layerfeatures. In this paper, the application of CNNs to one of the important soil-structureinteraction (SSI) problems, i.e., the analysis of vibrations transmission from the free-field next to a building to the building foundation, is presented in the case of mine-induced vibrations. To achieve this, the dataset from in-situ experimental measurements,containing 1D ground acceleration records, was converted into 2D spectrogram imagesusing either Fourier transform or continuous wavelet transform. Next, these images wereused as input for a pre-trained CNN. The output is a ratio of maximal vibration valuesrecorded simultaneously on the building foundation and on the ground. Therefore, the lastlayer of the CNN had to be changed from a classification to a regression one. The obtainedresults indicate the suitability of CNN for the analyzed problem.
Źródło:
Computer Assisted Methods in Engineering and Science; 2024, 31, 1; 3-28
2299-3649
Pojawia się w:
Computer Assisted Methods in Engineering and Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A comparative study on performance of basic and ensemble classifiers with various datasets
Autorzy:
Gunakala, Archana
Shahid, Afzal Hussain
Powiązania:
https://bibliotekanauki.pl/articles/30148255.pdf
Data publikacji:
2023
Wydawca:
Polskie Towarzystwo Promocji Wiedzy
Tematy:
classification
Naïve Bayes
neural network
Support Vector Machine
Decision Tree
ensemble learning
Random Forest
Opis:
Classification plays a critical role in machine learning (ML) systems for processing images, text and high -dimensional data. Predicting class labels from training data is the primary goal of classification. An optimal model for a particular classification problem is chosen based on the model's performance and execution time. This paper compares and analyzes the performance of basic as well as ensemble classifiers utilizing 10-fold cross validation and also discusses their essential concepts, advantages, and disadvantages. In this study five basic classifiers namely Naïve Bayes (NB), Multi-layer Perceptron (MLP), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF) and the ensemble of all the five classifiers along with few more combinations are compared with five University of California Irvine (UCI) ML Repository datasets and a Diabetes Health Indicators dataset from Kaggle repository. To analyze and compare the performance of classifiers, evaluation metrics like Accuracy, Recall, Precision, Area Under Curve (AUC) and F-Score are used. Experimental results showed that SVM performs best on two out of the six datasets (Diabetes Health Indicators and waveform), RF performs best for Arrhythmia, Sonar, Tic-tac-toe datasets, and the best ensemble combination is found to be DT+SVM+RF on Ionosphere dataset having respective accuracies 72.58%, 90.38%, 81.63%, 73.59%, 94.78% and 94.01%. The proposed ensemble combinations outperformed the conven¬tional models for few datasets.
Źródło:
Applied Computer Science; 2023, 19, 1; 107-132
1895-3735
2353-6977
Pojawia się w:
Applied Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A few-shot fine-grained image recognition method
Autorzy:
Wang, Jianwei
Chen, Deyun
Powiązania:
https://bibliotekanauki.pl/articles/2204540.pdf
Data publikacji:
2023
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
few-shot learning
attention metric
CNN
convolutional neural network
feature expression
wskaźnik uwagi
sieć neuronowa splotowa
cechy wyrażeń
Opis:
Deep learning methods benefit from data sets with comprehensive coverage (e.g., ImageNet, COCO, etc.), which can be regarded as a description of the distribution of real-world data. The models trained on these datasets are considered to be able to extract general features and migrate to a domain not seen in downstream. However, in the open scene, the labeled data of the target data set are often insufficient. The depth models trained under a small amount of sample data have poor generalization ability. The identification of new categories or categories with a very small amount of sample data is still a challenging task. This paper proposes a few-shot fine-grained image recognition method. Feature maps are extracted by a CNN module with an embedded attention network to emphasize the discriminative features. A channel-based feature expression is applied to the base class and novel class followed by an improved cosine similarity-based measurement method to get the similarity score to realize the classification. Experiments are performed on main few-shot benchmark datasets to verify the efficiency and generality of our model, such as Stanford Dogs, CUB-200, and so on. The experimental results show that our method has more advanced performance on fine-grained datasets.
Źródło:
Bulletin of the Polish Academy of Sciences. Technical Sciences; 2023, 71, 1; art. no. e144584
0239-7528
Pojawia się w:
Bulletin of the Polish Academy of Sciences. Technical Sciences
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A genetic algorithm based optimized convolutional neural network for face recognition
Autorzy:
Karlupia, Namrata
Mahajan, Palak
Abrol, Pawanesh
Lehana, Parveen K.
Powiązania:
https://bibliotekanauki.pl/articles/2201023.pdf
Data publikacji:
2023
Wydawca:
Uniwersytet Zielonogórski. Oficyna Wydawnicza
Tematy:
convolutional neural network
genetic algorithm
deep learning
evolutionary technique
sieć neuronowa konwolucyjna
algorytm genetyczny
uczenie głębokie
technika ewolucyjna
Opis:
Face recognition (FR) is one of the most active research areas in the field of computer vision. Convolutional neural networks (CNNs) have been extensively used in this field due to their good efficiency. Thus, it is important to find the best CNN parameters for its best performance. Hyperparameter optimization is one of the various techniques for increasing the performance of CNN models. Since manual tuning of hyperparameters is a tedious and time-consuming task, population based metaheuristic techniques can be used for the automatic hyperparameter optimization of CNNs. Automatic tuning of parameters reduces manual efforts and improves the efficiency of the CNN model. In the proposed work, genetic algorithm (GA) based hyperparameter optimization of CNNs is applied for face recognition. GAs are used for the optimization of various hyperparameters like filter size as well as the number of filters and of hidden layers. For analysis, a benchmark dataset for FR with ninety subjects is used. The experimental results indicate that the proposed GA-CNN model generates an improved model accuracy in comparison with existing CNN models. In each iteration, the GA minimizes the objective function by selecting the best combination set of CNN hyperparameters. An improved accuracy of 94.5% is obtained for FR.
Źródło:
International Journal of Applied Mathematics and Computer Science; 2023, 33, 1; 21--31
1641-876X
2083-8492
Pojawia się w:
International Journal of Applied Mathematics and Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A novel method for automatic detection of arrhythmias using the unsupervised convolutional neural network
Autorzy:
Zhang, Junming
Yao, Ruxian
Gao, Jinfeng
Li, Gangqiang
Wu, Haitao
Powiązania:
https://bibliotekanauki.pl/articles/23944827.pdf
Data publikacji:
2023
Wydawca:
Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
Tematy:
convolutional neural network
arrhythmia detection
unsupervised learning
ECG classification
Opis:
In recent years, various models based on convolutional neural networks (CNN) have been proposed to solve the cardiac arrhythmia detection problem and achieved saturated accuracy. However, these models are often viewed as “blackbox” and lack of interpretability, which hinders the understanding of cardiologists, and ultimately hinders the clinical use of intelligent terminals. At the same time, most of these approaches are supervised learning and require label data. It is a time-consuming and expensive process to obtain label data. Furthermore, in human visual cortex, the importance of lateral connection is same as feed-forward connection. Until now, CNN based on lateral connection have not been studied thus far. Consequently, in this paper, we combines CNNs, lateral connection and autoencoder (AE) to propose the building blocks of lateral connection convolutional autoencoder neural networks (LCAN) for cardiac arrhythmia detection, which learn representations in an unsupervised manner. Concretely, the LCAN contains a convolution layer, a lateral connection layer, an AE layer, and a pooling layer. The LCAN detects salient wave features through the lateral connection layer. The AE layer and competitive learning is used to update the filters of the convolution network—an unsupervised process that ensures similar weight distribution for all adjacent filters in each convolution layer and realizes the neurons’ semantic arrangement in the LCAN. To evaluate the performances of the proposed model, we have implemented the experiments on the well-known MIT–BIH Arrhythmia Database. The proposed model yields total accuracies and kappa coefficients of 98% and 0.95, respectively. The experiment results show that the LCAN is not only effective, but also a useful tool for arrhythmia detection.
Źródło:
Journal of Artificial Intelligence and Soft Computing Research; 2023, 13, 3; 181--196
2083-2567
2449-6499
Pojawia się w:
Journal of Artificial Intelligence and Soft Computing Research
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Adaptive Rider Feedback Artificial Tree Optimization-Based Deep Neuro-Fuzzy Network for Classification of Sentiment Grade
Autorzy:
Jasti, Sireesha
Kumar, G.V.S. Raj
Powiązania:
https://bibliotekanauki.pl/articles/2200961.pdf
Data publikacji:
2023
Wydawca:
Instytut Łączności - Państwowy Instytut Badawczy
Tematy:
deep learning network
feedback artificial tree
natural language processing (NLP)
rider optimization algorithm
sentiment grade classification
Opis:
Sentiment analysis is an efficient technique for expressing users’ opinions (neutral, negative or positive) regarding specific services or products. One of the important benefits of analyzing sentiment is in appraising the comments that users provide or service providers or services. In this work, a solution known as adaptive rider feedback artificial tree optimization-based deep neuro-fuzzy network (RFATO-based DNFN) is implemented for efficient sentiment grade classification. Here, the input is pre-processed by employing the process of stemming and stop word removal. Then, important factors, e.g. SentiWordNet-based features, such as the mean value, variance, as well as kurtosis, spam word-based features, term frequency-inverse document frequency (TF-IDF) features and emoticon-based features, are extracted. In addition, angular similarity and the decision tree model are employed for grouping the reviewed data into specific sets. Next, the deep neuro-fuzzy network (DNFN) classifier is used to classify the sentiment grade. The proposed adaptive rider feedback artificial tree optimization (A-RFATO) approach is utilized for the training of DNFN. The A-RFATO technique is a combination of the feedback artificial tree (FAT) approach and the rider optimization algorithm (ROA) with an adaptive concept. The effectiveness of the proposed A-RFATO-based DNFN model is evaluated based on such metrics as sensitivity, accuracy, specificity, and precision. The sentiment grade classification method developed achieves better sensitivity, accuracy, specificity, and precision rates when compared with existing approaches based on Large Movie Review Dataset, Datafiniti Product Database, and Amazon reviews.
Źródło:
Journal of Telecommunications and Information Technology; 2023, 1; 37--50
1509-4553
1899-8852
Pojawia się w:
Journal of Telecommunications and Information Technology
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
An investigation of the relationship between encoder difference and thermo-elastic machine tool deformation
Autorzy:
Brecher, Christian
Dehn, Mathias
Neus, Stephan
Powiązania:
https://bibliotekanauki.pl/articles/24084708.pdf
Data publikacji:
2023
Wydawca:
Wrocławska Rada Federacji Stowarzyszeń Naukowo-Technicznych
Tematy:
machine tool
thermal error compensation
machine learning
artificial neural network
Opis:
New approaches, using machine learning to model the thermo-elastic machine tool error, often rely on machine internal data, like axis speed or axis position as input data, which have a delayed relation to the thermo-elastic error. Since there is no direct relation to the thermo-elastic error, this can lead to an increased computation inaccuracy of the model or the need for expensive sensor equipment for additional input data. The encoder difference is easy to obtain and has a direct relationship with the thermo-elastic error and therefore has a high potential to improve the accuracy thermo-elastic error models. This paper first investigates causes of the encoder difference and its relationship with the thermo-elastic error. Afterwards, the model is presented, which uses the encoder difference to compute the thermo-elastic error. Due to the complexity of the relationship, it is necessary, to use a machine learning approach for this. To conclude, the potential of the encoder difference as an input of the model is evaluated.
Źródło:
Journal of Machine Engineering; 2023, 23, 3; 26--37
1895-7595
2391-8071
Pojawia się w:
Journal of Machine Engineering
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Anonymous traffic classification based on three-dimensional Markov image and deep learning
Autorzy:
Tang, Xin
Li, Huanzhou
Zhang, Jian
Tang, Zhangguo
Wang, Han
Cai, Cheng
Powiązania:
https://bibliotekanauki.pl/articles/27311448.pdf
Data publikacji:
2023
Wydawca:
Polska Akademia Nauk. Czasopisma i Monografie PAN
Tematy:
anonymous network
traffic classification
three-dimensional Markov image
output self-attention
deep learning
sieć anonimowa
klasyfikacja ruchu
trójwymiarowy obraz Markowa
samouwaga wyjściowa
uczenie głębokie
Opis:
Illegal elements use the characteristics of an anonymous network hidden service mechanism to build a dark network and conduct various illegal activities, which brings a serious challenge to network security. The existing anonymous traffic classification methods suffer from cumbersome feature selection and difficult feature information extraction, resulting in low accuracy of classification. To solve this problem, a classification method based on three-dimensional Markov images and output self-attention convolutional neural network is proposed. This method first divides and cleans anonymous traffic data packets according to sessions, then converts the cleaned traffic data into three-dimensional Markov images according to the transition probability matrix of bytes, and finally inputs the images to the output self-attention convolution neural network to train the model and perform classification. The experimental results show that the classification accuracy and F1-score of the proposed method for Tor, I2P, Freenet, and ZeroNet can exceed 98.5%, and the average classification accuracy and F1-score for 8 kinds of user behaviors of each type of anonymous traffic can reach 93.7%. The proposed method significantly improves the classification effect of anonymous traffic compared with the existing methods.
Źródło:
Bulletin of the Polish Academy of Sciences. Technical Sciences; 2023, 71, 4; art. no. e145676
0239-7528
Pojawia się w:
Bulletin of the Polish Academy of Sciences. Technical Sciences
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Artificial intelligence applications in project scheduling: a systematic review, bibliometric analysis, and prospects for future research
Autorzy:
Bahroun, Zied
Tanash, Moayad
Ad, Rami As
Alnajar, Mohamad
Powiązania:
https://bibliotekanauki.pl/articles/27315576.pdf
Data publikacji:
2023
Wydawca:
STE GROUP
Tematy:
artificial intelligence
machine learning
project scheduling
bibliometric analysis
network analysis
review
Opis:
The availability of digital infrastructures and the fast-paced development of accompanying revolutionary technologies have triggered an unprecedented reliance on Artificial intelligence (AI) techniques both in theory and practice. Within the AI domain, Machine Learning (ML) techniques stand out as essential facilitator largely enabling machines to possess human-like cognitive and decision making capabilities. This paper provides a focused review of the literature addressing applications of emerging ML toolsto solve various Project Scheduling Problems (PSPs). In particular, it employs bibliometric and network analysis tools along with a systematic literature review to analyze a pool of 104 papers published between 1985 and August 2021. The conducted analysis unveiled the top contributing authors, the most influential papers as well as the existing research tendencies and thematic research topics within this field of study. A noticeable growth in the number of relevant studies is seen recently with a steady increase as of the year 2018. Most of the studies adopted Artificial Neural Networks, Bayesian Network and Reinforcement Learning techniques to tackle PSPs under a stochastic environment, where these techniques are frequently hybridized with classical metaheuristics. The majority of works (57%) addressed basic Resource Constrained PSPs and only 15% are devoted to the project portfolio management problem. Furthermore, this study clearly indicates that the application of AI techniques to efficiently handle PSPs is still in its infancy stage bringing out the need for further research in this area. This work also identifies current research gaps and highlights a multitude of promising avenues for future research.
Źródło:
Management Systems in Production Engineering; 2023, 2 (31); 144--161
2299-0461
Pojawia się w:
Management Systems in Production Engineering
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Artificial neural network (ANN) modelling to estimate bubble size from macroscopic image and object features
Autorzy:
Vinnett, Luis
León, Roberto
Mesa, Diego
Powiązania:
https://bibliotekanauki.pl/articles/29552038.pdf
Data publikacji:
2023
Wydawca:
Politechnika Wrocławska. Oficyna Wydawnicza Politechniki Wrocławskiej
Tematy:
machine learning
artificial neural network
flotation
bubble size
Sauter diameter
Opis:
Bubble size measurements in aerated systems such as froth flotation cells are critical for controlling gas dispersion. Commonly, bubbles are measured by obtaining representative photographs, which are then analyzed using segmentation and identification software tools. Recent developments have focused on enhancing these segmentation tools. However, the main challenges around complex bubble cluster segmentation remain unresolved, while the tools to tackle these challenges have become increasingly complex and computationally expensive. In this work, we propose an alternative solution, circumventing the need for image segmentation and bubble identification. An Artificial Neural Network (ANN) was trained to estimate the Sauter mean bubble size (D32) based on macroscopic image features obtained with simple and inexpensive image analysis. The results showed excellent prediction accuracy, with a correlation coefficient, R, over 0.998 in the testing stage, and without bias in its error distribution. This machine learning tool paves the way for robust and fast estimation of bubble size under complex bubble images, without the need of image segmentation.
Źródło:
Physicochemical Problems of Mineral Processing; 2023, 59, 5; art. no. 185759
1643-1049
2084-4735
Pojawia się w:
Physicochemical Problems of Mineral Processing
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Choice of the p-norm for high level classification features pruning in modern convolutional neural networks with local sensitivity analysis
Autorzy:
Jeczmionek, Ernest
Kowalski, Piotr A.
Powiązania:
https://bibliotekanauki.pl/articles/24988509.pdf
Data publikacji:
2023
Wydawca:
Uniwersytet Zielonogórski. Oficyna Wydawnicza
Tematy:
convolutional neural network
pruning
sensitivity analysis
transfer learning
ImageNet
sieć neuronowa konwolucyjna
analiza wrażliwości
uczenie transferowe
Opis:
Transfer learning has surfaced as a compelling technique in machine learning, enabling the transfer of knowledge across networks. This study evaluates the efficacy of ImageNet pretrained state-of-the-art networks, including DenseNet, ResNet, and VGG, in implementing transfer learning for prepruned models on compact datasets, such as FashionMNIST, CIFAR10, and CIFAR100. The primary objective is to reduce the number of neurons while preserving high-level features. To this end, local sensitivity analysis is employed alongside p-norms and various reduction levels. This investigation discovers that VGG16, a network rich in parameters, displays resilience to high-level feature pruning. Conversely, the ResNet architectures reveal an interesting pattern of increased volatility. These observations assist in identifying an optimal combination of the norm and the reduction level for each network architecture, thus offering valuable directions for model-specific optimization. This study marks a significant advance in understanding and implementing effective pruning strategies across diverse network architectures, paving the way for future research and applications.
Źródło:
International Journal of Applied Mathematics and Computer Science; 2023, 33, 4; 663--672
1641-876X
2083-8492
Pojawia się w:
International Journal of Applied Mathematics and Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Classification of Parkinsons disease in brain MRI images using Deep Residual Convolutional Neural Network (DRCNN)
Autorzy:
Praneeth, Puppala
Sathvika, Majety
Kommareddy, Vivek
Sarath, Madala
Mallela, Saran
Vani, K. Suvarna
Chkrabarti, Prasun
Powiązania:
https://bibliotekanauki.pl/articles/30148251.pdf
Data publikacji:
2023
Wydawca:
Polskie Towarzystwo Promocji Wiedzy
Tematy:
Parkinson’s disease
Deep Residual Convolutional Neural Network
deep learning
health control
Opis:
In our aging culture, neurodegenerative disorders like Parkinson's disease (PD) are among the most serious health issues. It is a neurological condition that has social and economic effects on individuals. It happens because the brain's dopamine-producing cells are unable to produce enough of the chemical to support the body's motor functions. The main symptoms of this illness are eyesight, excretion activity, speech, and mobility issues, followed by depression, anxiety, sleep issues, and panic attacks. The main aim of this research is to develop a workable clinical decision-making framework that aids the physician in diagnosing patients with PD influence. In this research, the authors propose a technique to classify Parkinson’s disease by MRI brain images. Initially, the input data is normalized using the min-max normalization method, and then noise is removed from the input images using a median filter. The Binary Dragonfly algorithm is then used to select features. In addition, the Dense-UNet technique is used to segment the diseased part from brain MRI images. The disease is then classified as Parkinson's disease or health control using the Deep Residual Convolutional Neural Network (DRCNN) technique along with the Enhanced Whale Optimization Algorithm (EWOA) to achieve better classification accuracy. In this work, the Parkinson's Progression Marker Initiative (PPMI) public dataset for Parkinson's MRI images is used. Indicators of accuracy, sensitivity, specificity and precision are used with manually collected data to evaluate the effectiveness of the proposed methodology.
Źródło:
Applied Computer Science; 2023, 19, 2; 125-146
1895-3735
2353-6977
Pojawia się w:
Applied Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Constant Q-transform-based deep learning architecture for detection of obstructive sleep apnea
Autorzy:
Kandukuri, Usha Rani
Prakash, Allam Jaya
Patro, Kiran Kumar
Neelapu, Bala Chakravarthy
Tadeusiewicz, Ryszard
Pławiak, Paweł
Powiązania:
https://bibliotekanauki.pl/articles/24200694.pdf
Data publikacji:
2023
Wydawca:
Uniwersytet Zielonogórski. Oficyna Wydawnicza
Tematy:
sleep apnea
convolutional neural network
constant Q-transform
deep learning
single lead ECG signal
non apnea
obstructive sleep apnea
bezdech senny
sieć neuronowa konwolucyjna
uczenie głębokie
sygnał EKG
obturacyjny bezdech senny
Opis:
Obstructive sleep apnea (OSA) is a long-term sleep disorder that causes temporary disruption in breathing while sleeping. Polysomnography (PSG) is the technique for monitoring different signals during the patient’s sleep cycle, including electroencephalogram (EEG), electromyography (EMG), electrocardiogram (ECG), and oxygen saturation (SpO2). Due to the high cost and inconvenience of polysomnography, the usefulness of ECG signals in detecting OSA is explored in this work, which proposes a two-dimensional convolutional neural network (2D-CNN) model for detecting OSA using ECG signals. A publicly available apnea ECG database from PhysioNet is used for experimentation. Further, a constant Q-transform (CQT) is applied for segmentation, filtering, and conversion of ECG beats into images. The proposed CNN model demonstrates an average accuracy, sensitivity and specificity of 91.34%, 90.68% and 90.70%, respectively. The findings obtained using the proposed approach are comparable to those of many other existing methods for automatic detection of OSA.
Źródło:
International Journal of Applied Mathematics and Computer Science; 2023, 33, 3; 493--506
1641-876X
2083-8492
Pojawia się w:
International Journal of Applied Mathematics and Computer Science
Dostawca treści:
Biblioteka Nauki
Artykuł

Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies