Influence of data scaling and normalization on overall neural network performances in photoacoustics
Само за регистроване кориснике
2022
Аутори
Đorđević, Katarina Lj.Jordović-Pavlović, Miroslava I.
Ćojbašić, Ž. M.
Galović, Slobodanka
Popović, Marica N.
Nešić, Mioljub V.
Markushev, Dragan D.
Чланак у часопису (Објављена верзија)
Метаподаци
Приказ свих података о документуАпстракт
In this paper, the influence of the input and output data scaling and normalization on the neural network overall performances is investigated aimed at inverse problem-solving in photoacoustics of semiconductors. The logarithmic scaling of the photoacoustic signal amplitudes as input data and numerical scaling of the sample thermal parameters as output data are presented as useful tools trying to reach maximal network precision. Max and min–max normalizations to the input data are presented to change their numerical values in the dataset to common scales, without distorting differences. It was demonstrated in theory that the largest network prediction error of all targeted parameters is obtained by a network with non-scaled output data. Also, it was found out that the best network prediction was achieved with min–max normalization of the input data and network predicted output data scale within the range of [1–10]. Network training and prediction performances analyzed with experimental... input data show that the benefits and improvements of input and output scaling and normalization are not guaranteed but are strongly dependent on a specific problem to be solved.
Кључне речи:
Artificial neural networks / Inverse problem / Photoacoustic / Photothermal / Semiconductors / Thermal diffusion / Thermal expansionИзвор:
Optical and Quantum Electronics, 2022, 54, 8, 501-Финансирање / пројекти:
- Министарство науке, технолошког развоја и иновација Републике Србије, институционално финансирање - 200017 (Универзитет у Београду, Институт за нуклеарне науке Винча, Београд-Винча) (RS-MESTD-inst-2020-200017)
DOI: 10.1007/s11082-022-03799-1
ISSN: 1572-817X
WoS: 00082493590000
Scopus: 2-s2.0-85134068645
Колекције
Институција/група
VinčaTY - JOUR AU - Đorđević, Katarina Lj. AU - Jordović-Pavlović, Miroslava I. AU - Ćojbašić, Ž. M. AU - Galović, Slobodanka AU - Popović, Marica N. AU - Nešić, Mioljub V. AU - Markushev, Dragan D. PY - 2022 UR - https://vinar.vin.bg.ac.rs/handle/123456789/10359 AB - In this paper, the influence of the input and output data scaling and normalization on the neural network overall performances is investigated aimed at inverse problem-solving in photoacoustics of semiconductors. The logarithmic scaling of the photoacoustic signal amplitudes as input data and numerical scaling of the sample thermal parameters as output data are presented as useful tools trying to reach maximal network precision. Max and min–max normalizations to the input data are presented to change their numerical values in the dataset to common scales, without distorting differences. It was demonstrated in theory that the largest network prediction error of all targeted parameters is obtained by a network with non-scaled output data. Also, it was found out that the best network prediction was achieved with min–max normalization of the input data and network predicted output data scale within the range of [1–10]. Network training and prediction performances analyzed with experimental input data show that the benefits and improvements of input and output scaling and normalization are not guaranteed but are strongly dependent on a specific problem to be solved. T2 - Optical and Quantum Electronics T1 - Influence of data scaling and normalization on overall neural network performances in photoacoustics VL - 54 IS - 8 SP - 501 DO - 10.1007/s11082-022-03799-1 ER -
@article{ author = "Đorđević, Katarina Lj. and Jordović-Pavlović, Miroslava I. and Ćojbašić, Ž. M. and Galović, Slobodanka and Popović, Marica N. and Nešić, Mioljub V. and Markushev, Dragan D.", year = "2022", abstract = "In this paper, the influence of the input and output data scaling and normalization on the neural network overall performances is investigated aimed at inverse problem-solving in photoacoustics of semiconductors. The logarithmic scaling of the photoacoustic signal amplitudes as input data and numerical scaling of the sample thermal parameters as output data are presented as useful tools trying to reach maximal network precision. Max and min–max normalizations to the input data are presented to change their numerical values in the dataset to common scales, without distorting differences. It was demonstrated in theory that the largest network prediction error of all targeted parameters is obtained by a network with non-scaled output data. Also, it was found out that the best network prediction was achieved with min–max normalization of the input data and network predicted output data scale within the range of [1–10]. Network training and prediction performances analyzed with experimental input data show that the benefits and improvements of input and output scaling and normalization are not guaranteed but are strongly dependent on a specific problem to be solved.", journal = "Optical and Quantum Electronics", title = "Influence of data scaling and normalization on overall neural network performances in photoacoustics", volume = "54", number = "8", pages = "501", doi = "10.1007/s11082-022-03799-1" }
Đorđević, K. Lj., Jordović-Pavlović, M. I., Ćojbašić, Ž. M., Galović, S., Popović, M. N., Nešić, M. V.,& Markushev, D. D.. (2022). Influence of data scaling and normalization on overall neural network performances in photoacoustics. in Optical and Quantum Electronics, 54(8), 501. https://doi.org/10.1007/s11082-022-03799-1
Đorđević KL, Jordović-Pavlović MI, Ćojbašić ŽM, Galović S, Popović MN, Nešić MV, Markushev DD. Influence of data scaling and normalization on overall neural network performances in photoacoustics. in Optical and Quantum Electronics. 2022;54(8):501. doi:10.1007/s11082-022-03799-1 .
Đorđević, Katarina Lj., Jordović-Pavlović, Miroslava I., Ćojbašić, Ž. M., Galović, Slobodanka, Popović, Marica N., Nešić, Mioljub V., Markushev, Dragan D., "Influence of data scaling and normalization on overall neural network performances in photoacoustics" in Optical and Quantum Electronics, 54, no. 8 (2022):501, https://doi.org/10.1007/s11082-022-03799-1 . .