An approach to evaluation of the extremely low-frequency magnetic field radiation in the laptop computer neighborhood by artificial neural networks

• Published in: Neural computing & applications Vol. 28; no. 11; pp. 3441 - 3453
• Main Authors: Brodić, Darko, Tanikić, Dejan, Amelio, Alessia
• Format: Journal Article
• Language: English
• Published: London Springer London 01.11.2017; Springer Nature B.V
• Subjects: Artificial Intelligence; Artificial neural networks; Clusters; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer industry; Computer Science; Computer simulation; Correlation coefficients; Data Mining and Knowledge Discovery; Image Processing and Computer Vision; Magnetic fields; Neural networks; Original Article; Portable computers; Predictions; Probability and Statistics in Computer Science; Self organizing maps; Portable computers; Self-organizing map; Artificial neural networks; Classification; Non-ionizing radiation; Exposure; Magnetic field
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-016-2246-3

The paper considers the level of the extremely low-frequency magnetic field produced by laptop computers, which can be hazardous to the laptop users’ health. It proposes a new model for prediction of the laptop magnetic field radiation starting from a priori known laptop characteristics using artificial neural network. In the experiment, the magnetic field of ten different laptop computers is measured. The measuring of the magnetic field is performed at the points, which are relevant to laptop computer users. Prediction is obtained from testing laptops. The predicted magnetic field results are evaluated by correlation coefficient and clustered for detection of magnetic field ranges by self-organizing map. Obtained cluster set is compared with the cluster set computed on the measured magnetic field at the same laptop characteristics. Correlation coefficient showed good correspondence of the predicted values with the measured values. Furthermore, the comparison between the two cluster sets revealed a good alignment of magnetic field ranges in measured and predicted results.