Classification of Solo Batik patterns using deep learning convolutional neural networks algorithm DOI Creative Commons
Dimas Aryo Anggoro,

Assyati Amadjida Tamimi Marzuki,

Wiwit Supriyanti

и другие.

TELKOMNIKA (Telecommunication Computing Electronics and Control), Год журнала: 2023, Номер 22(1), С. 232 - 232

Опубликована: Авг. 30, 2023

The ideology of the Solo Batik pattern has not been conveyed to public. In addition, a lot people are unaware that batik contains particular patterns also used for activities. This study uses convolutional neural network model categorize 9 different according their use elaborate geometric shapes, complicated symbols, patterns, dots, and natural designs. With 1 4 hidden layers, we aim select number layers yields highest accuracy. A 100×100 pixel image is as input. feature extraction process then makes 3×3 maps from three convolution layers. dropout regularization added, with settings ranging 0.1 0.9. Adam algorithm in this perform optimization. 3-layered networks (CNN) value 0.2, run 20 epochs, produced accuracy results 97.77%, which was highest. Additionally, it can be inferred applying certain adding right values an impact on raising score.

Язык: Английский

Machine learning techniques in bankruptcy prediction: A systematic literature review DOI
Απόστολος Δασίλας,

Anna Rigani

Expert Systems with Applications, Год журнала: 2024, Номер 255, С. 124761 - 124761

Опубликована: Июль 14, 2024

Язык: Английский

Процитировано

9
RETRACTED ARTICLE: Explainable AI Model for Recognizing Financial Crisis Roots Based on Pigeon Optimization and Gradient Boosting Model DOI Creative Commons
Mohamed Torky, Ibrahim Gad, Aboul Ella Hassanien

и другие.

International Journal of Computational Intelligence Systems, Год журнала: 2023, Номер 16(1)

Опубликована: Апрель 5, 2023

Abstract Utilizing Artificial Intelligence (AI) techniques to forecast, recognize, and classify financial crisis roots are important research challenges that have attracted the interest of researchers. Moreover, Explainable (XAI) concept enables AI interpret results processing testing complex data patterns so humans can find efficient ways infer logic behind classifying patterns. This paper proposes a novel XAI model automatically recognize interprets features selection operation. Using benchmark dataset, proposed utilized pigeon optimizer optimize feature operation, then Gradient Boosting classifier is based on obtained reduct most features. The practical showed short-term rates by which be detected. classification built-in in Pigeon Inspired Optimizer (PIO) algorithm achieved training accuracy 99% 96.7%, respectively, recognizing roots, an better performance compared random forest classifier.

Язык: Английский

Процитировано

16
Classification of Solo Batik patterns using deep learning convolutional neural networks algorithm DOI Creative Commons
Dimas Aryo Anggoro,

Assyati Amadjida Tamimi Marzuki,

Wiwit Supriyanti

и другие.

TELKOMNIKA (Telecommunication Computing Electronics and Control), Год журнала: 2023, Номер 22(1), С. 232 - 232

Опубликована: Авг. 30, 2023

The ideology of the Solo Batik pattern has not been conveyed to public. In addition, a lot people are unaware that batik contains particular patterns also used for activities. This study uses convolutional neural network model categorize 9 different according their use elaborate geometric shapes, complicated symbols, patterns, dots, and natural designs. With 1 4 hidden layers, we aim select number layers yields highest accuracy. A 100×100 pixel image is as input. feature extraction process then makes 3×3 maps from three convolution layers. dropout regularization added, with settings ranging 0.1 0.9. Adam algorithm in this perform optimization. 3-layered networks (CNN) value 0.2, run 20 epochs, produced accuracy results 97.77%, which was highest. Additionally, it can be inferred applying certain adding right values an impact on raising score.

Язык: Английский

Процитировано

1