Advanced search
Publication Cover
Cogent Engineering Volume 7, 2020 - Issue 1
Submit an article Journal homepage
3,565
Views
9
CrossRef citations to date
0
Altmetric
COMPUTER SCIENCE

One method of generating synthetic data to assess the upper limit of machine learning algorithms performance

Yan I. Kuchin1 Institute of Cybernetics and Information Technology, Satbayev University, Almaty, 050013, The Republic of Kazakhstan;2 Laboratory of analysis and modeling of information processes, Institute of Information and Computational Technologies, Almaty, 050010, Kazakhstan;3 Department of Artificial Intelligence and Systems Engineering, Riga Technical University, Riga, LV, 1658, LatviaCorrespondence[email protected]
ORCID Iconhttps://orcid.org/0000-0002-5271-9071View further author information
ORCID Icon,
Ravil I. Mukhamediev2 Laboratory of analysis and modeling of information processes, Institute of Information and Computational Technologies, Almaty, 050010, Kazakhstan;3 Department of Artificial Intelligence and Systems Engineering, Riga Technical University, Riga, LV, 1658, LatviaORCID Iconhttps://orcid.org/0000-0002-3727-043XView further author information
ORCID Icon &
Kirill O. Yakunin2 Laboratory of analysis and modeling of information processes, Institute of Information and Computational Technologies, Almaty, 050010, Kazakhstan;3 Department of Artificial Intelligence and Systems Engineering, Riga Technical University, Riga, LV, 1658, LatviaORCID Iconhttps://orcid.org/0000-0002-7378-9212View further author information
ORCID Icon |
Duc Pham4 School of Mechanical Engineering, University of Birmingham, Birmingham, United KingdomView further author information
(Reviewing editor)
Article: 1718821 | Received 30 Jul 2019, Accepted 12 Jan 2020, Published online: 03 Feb 2020

Figures & data

Figure 1. Distribution of AR values for 13 lithotypes

(Each lithotype is color/number coded).
Figure 1. Distribution of AR values for 13 lithotypes

Figure 2. “Measured” and generated AR values

Figure 2. “Measured” and generated AR values

Figure 3. Modeled (left column) and actual (right column) log-data for AR (blue) and SP (green) values

Figure 3. Modeled (left column) and actual (right column) log-data for AR (blue) and SP (green) values

Figure 4. LSTM memory cell (Hochreiter & Schmidhuber, Citation1997)

Figure 4. LSTM memory cell (Hochreiter & Schmidhuber, Citation1997)

Figure 5. LSTM network architecture implemented in Keras

Figure 5. LSTM network architecture implemented in Keras

Figure 6. Floating data window

Figure 6. Floating data window

Table 1. Results of classifiers on simulated data

Table

Table 2. Results of classifiers on real data

Table

Table 3. Classification results for partial replacement of real data with simulated data

Table 4. Classification results on a mixed dataset

Table 5. Classification results for different sizes of a floating window

Table

Table 6. Classification results when adding boreholes coordinates

Related Research Data

Integration of Results from Recognition Algorithms Applied to the Uranium Deposits
Source: Fuji Technology Press Ltd.
Neural networks in geophysical applications
Source: Society of Exploration Geophysicists
Lithology from downhole logs
Source: Oxford University Press (OUP)
Estimation of missing logs by regularized neural networks
Source: American Association of Petroleum Geologists AAPG/Datapages
Support-vector networks
Source: Springer Science and Business Media LLC
Original Contribution: A scaled conjugate gradient algorithm for fast supervised learning
Source: Elsevier BV
The incidence of trigger finger or de Quervain’s tendinitis after carpal tunnel release:
Source: SAGE Publications
First Report of Alternaria dianthicola Causing Leaf Blight on Withania somnifera from India.
Source: Scientific Societies
a review
Source: Association for Computing Machinery (ACM)
Greedy function approximation: A gradient boosting machine.
Source: Institute of Mathematical Statistics
The Distance-Weighted k-Nearest-Neighbor Rule
Source: Institute of Electrical and Electronics Engineers (IEEE)
Neural networks for classification: a survey
Source: Institute of Electrical and Electronics Engineers (IEEE)
Comparative analysis of classification algorithms
Source: IEEE
Automated lithology prediction from PGNAA and other geophysical logs
Source: Elsevier BV
Recognition of Rocks at Uranium Deposits by Using a Few Methods of Machine Learning
Source: Springer International Publishing
On the limited memory BFGS method for large scale optimization
Source: Springer Science and Business Media LLC
Geophysical research of boreholes: Artificial neural networks data analysis
Source: IEEE

Linking provided by 

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.