Crop Yield Forecasting Based on Echo State Network Tuned by Crayfish Optimization Algorithm DOI
Nebojša Bačanin, Luka Jovanovic, Marko Djordjevic

et al.

Published: March 15, 2024

Language: Английский

Mapping Roofing with Asbestos-Containing Material by Using Remote Sensing Imagery and Machine Learning-Based Image Classification: A State-of-the-Art Review DOI Open Access
Mohammad Abbasi, Sherif Mostafa,

Abel Silva Vieira

et al.

Sustainability, Journal Year: 2022, Volume and Issue: 14(13), P. 8068 - 8068

Published: July 1, 2022

Building roofing produced with asbestos-containing materials is a significant concern due to its detrimental health hazard implications. Efficiently locating asbestos essential proactively mitigate and manage potential risks from this legacy building material. Several studies utilised remote sensing imagery machine learning-based image classification methods for mapping roofs materials. However, there has not yet been critical review of conducted in order provide coherent guidance on the use different images processes. This paper critically reviews latest works identify challenges discuss possible solutions improving process. A peer addressing roof published 2012 2022 was synthesise evaluate input types methods. Then, process were identified, suggested address identified challenges. The results showed that hyperspectral traditional pixel-based classifiers caused large omission errors. Classifying very-high-resolution multispectral by adopting object-based improved accuracy ACM identification; however, non-optimal segmentation parameters, inadequate training data supervised methods, analyst subjectivity rule-based classifications reported as While only one study investigated convolutional neural networks mapping, other applications demonstrated promising using deep-learning-based models. suggests further utilising Mask R-CNN 3D-CNN conventional approaches developing end-to-end deep semantic models map

Language: Английский

Citations

21

Sine Cosine Algorithm for Simple recurrent neural network Tuning for Stock Market Prediction DOI
Luka Jovanovic,

Nemanja Milutinovic,

Masa Gajevic

et al.

2022 30th Telecommunications Forum (TELFOR), Journal Year: 2022, Volume and Issue: unknown

Published: Nov. 15, 2022

Deep artificial neural networks have recently gained popularity in the time series forecasting literature. Recurrent networks’ higher suitability for this type of problem is reason why network has been chosen over other deep approaches. Due to number parameters used simplicity these considerable. This characteristic makes recurrent highly suitable problems forecasting. Unfortunately, finding architecture each specific task NP-hard, therefore employment metaheuristics appropriate. Accordingly, research proposed paper tackles tuning simple by sine cosine algorithm stock market prediction. The method’s performance was compared with and validated against Nikkei exchange.

Language: Английский

Citations

20

Forecasting Bitcoin Price by Tuned Long Short Term Memory Model DOI Creative Commons
Aleksandar Petrović, Luka Jovanovic, Miodrag Živković

et al.

Published: Jan. 1, 2023

The interest for cryptocurrencies is high and hence this work focuses on providing a practical real-world application of the swarm metaheuristics long short term memory model (LSTM).The goal price forecasting which interesting due to volatility cryptocurrencies.The authors apply LSTM solution problem has been proven reap results with type problem.The further optimized by metaheuristic -arithmetic optimization algorithm (AOA).The was tested alongside familiar high-performing competitors use standard metrics mean absolute error (MAE), squared (MSE), percentage (MAPE), root (RMSE).These have used comparison between solutions, upon proposed obtained overall best performance that testifies improvement solution.

Language: Английский

Citations

13

Artificial Neural Network Tuning by Improved Sine Cosine Algorithm for HealthCare 4.0 DOI Creative Commons

Masa Gajevic,

Nemanja Milutinovic,

Jelena Krstovic

et al.

Published: Jan. 1, 2023

This paper explores classification of datasets for Healthcare 4.0 using artificial neural networks which are tuned by improved sine cosine algorithm (SCA).Healthcare themes include internet things (IoT), industrial IoT (IIoT), cognitive computing, intelligence, cloud fog edge and other industry procedures.Health issues identification critical since prompt treatment improves the quality life individuals affected.One most difficult challenges intelligence (AI) is selecting control parameters that appropriate situation at hand.This presents a metaheuristics-based method training network, utilizing SCA.

Language: Английский

Citations

12

Crop Yield Forecasting Based on Echo State Network Tuned by Crayfish Optimization Algorithm DOI
Nebojša Bačanin, Luka Jovanovic, Marko Djordjevic

et al.

Published: March 15, 2024

Language: Английский

Citations

4