臺灣博碩士論文加值系統

Classifying or predicting soil type for Compressed Earth Block (CEB) construction using machine learning model is an important technique to replace laboratory tests which are time and cost consuming. The previous study has established soil classification using Artificial Neural Network (ANN). Nonetheless, gradient-based learning on ANN face several issues like overfitting and trapped in local minima due to poor generalization performance. In the further development of the neural network model, Extreme Learning Machine (ELM) has been developed with faster learning speed and better generalization performance compared to ANN. However, ELM itself has some drawbacks in random input weights and heavy memory problems. This research proposed the model called Backpropagated Deep Restricted Boltzmann Extreme Learning Machine (BD-RBELM) which establish deep ELM network to solve the heavy memory problem and use Restricted Boltzmann Machine (RBM) to train input weight for solving random input weights problem. The model then used for the soil classification problem with the final performance of 95.144% accuracy and 95.120 F1-Score, which performed better compared with the other AI techniques.

Classifying or predicting soil type for Compressed Earth Block (CEB) construction using machine learning model is an important technique to replace laboratory tests which are time and cost consuming. The previous study has established soil classification using Artificial Neural Network (ANN). Nonetheless, gradient-based learning on ANN face several issues like overfitting and trapped in local minima due to poor generalization performance. In the further development of the neural network model, Extreme Learning Machine (ELM) has been developed with faster learning speed and better generalization performance compared to ANN. However, ELM itself has some drawbacks in random input weights and heavy memory problems. This research proposed the model called Backpropagated Deep Restricted Boltzmann Extreme Learning Machine (BD-RBELM) which establish deep ELM network to solve the heavy memory problem and use Restricted Boltzmann Machine (RBM) to train input weight for solving random input weights problem. The model then used for the soil classification problem with the final performance of 95.144% accuracy and 95.120 F1-Score, which performed better compared with the other AI techniques.

ABBREVIATIONS AND SYMBOLS vi
LIST OF FIGURES ix
LIST OF TABLES x
CHAPTER 1: INTRODUCTION 1
1.1 Background 1
1.2 Research Objective 4
1.3 Research Scope and Assumptions 4
1.4 Research Methodology 5
1.5 Research Outline 8
CHAPTER 2: LITERATURE REVIEW 9
2.1 Soil Classification for CEB 9
2.1.1 ASTM standard soil analysis 9
2.1.2 Field Soil Analysis 10
2.1.3 Soil Dataset Input-Output 12
2.2 Data Imbalance 14
2.3 SMOTE: Synthetic Minority Over-sampling Technique 15
2.4 Restricted Boltzmann Machine (RBM) 17
2.5 Extreme Learning Machine (ELM) 19
2.6 Backpropagated Deep Extreme Learning Machine (BP-D-ELM) 21
CHAPTER 3: METHODOLOGY 26
3.1 Data Balancing using SMOTE 27
3.2 Data Normalization 27
3.3 Data Train-Test Split 27
3.4 Backpropagated Deep Restricted Boltzmann Extreme Learning Machine 28
3.5 Performance Evaluation & Measurement Criteria 38
3.5.1 Accuracy 38
3.5.2 Precision 38
3.5.3 Recall 39
3.5.4 F1-Score 39
3.5.5 Average Performance Score 40
CHAPTER 4: RESULT & EVALUATION 42
4.1 Dataset Information 42
4.2 Model Comparisons 44
4.2.1 Model Parameters 45
4.2.2 Result Analysis 46
4.2.3 Statistical Test 51
4.3 Confusion Matrix Analysis 54
CHAPTER 5: CONCLUSION 60
5.1 Conclusion 60
5.2 Recommendation 61
REFERENCES 62

Adam, EA, and ARA Agib. 2001. 'Compressed stabilised earth block manufacture in Sudan', France, Paris: Printed by Graphoprint for UNESCO.
ASTM, D. 2010. 'Standard test methods for liquid limit, plastic limit, and plasticity index of soils', ASTM International.
ASTM, D, "422-63 2007. 'Standard test method for particle-size analysis of soil', Annual book of American Society for Testing and Material standards, West Conshohocken, United States.
ASTM, D, "2487-11 2011. 'Standard practice for classification of soils for engineering purposes (unified soil classification system) Annual Book of ASTM Standard', ASTM International, 4: 08.
Castellanos, Francisco J., Jose J. Valero-Mas, Jorge Calvo-Zaragoza, and Juan R. Rico-Juan. 2018. 'Oversampling imbalanced data in the string space', Pattern Recognition Letters, 103: 32-38.
Chawla, Nitesh V., Kevin W. Bowyer, Lawrence O. Hall, and W. Philip Kegelmeyer. 2002. 'SMOTE: synthetic minority over-sampling technique', Journal of artificial intelligence research, 16: 321-57.
Douzas, Georgios, Fernando Bacao, and Felix Last. 2018. 'Improving imbalanced learning through a heuristic oversampling method based on k-means and SMOTE', Information Sciences, 465: 1-20.
Egenti, C., J. M. Khatib, and D. Oloke. 2014. 'Conceptualisation and pilot study of shelled compressed earth block for sustainable housing in Nigeria', International Journal of Sustainable Built Environment, 3: 72-86.
Erhan, Dumitru, Yoshua Bengio, Aaron Courville, Pierre-Antoine Manzagol, Pascal Vincent, and Samy Bengio. 2010. 'Why does unsupervised pre-training help deep learning?', Journal of Machine Learning Research, 11: 625-60.
Fan, Jiancong, Zhonghan Niu, Yongquan Liang, and Zhongying Zhao. 2016. 'Probability model selection and parameter evolutionary estimation for clustering imbalanced data without sampling', Neurocomputing, 211: 172-81.
Fotouhi, Sara, Shahrokh Asadi, and Michael W. Kattan. 2019. 'A comprehensive data level analysis for cancer diagnosis on imbalanced data', Journal of Biomedical Informatics, 90: 103089.
Fu, Aimin, Chunru Dong, and Laisheng Wang. 2015. 'An experimental study on stability and generalization of extreme learning machines', International Journal of Machine Learning and Cybernetics, 6: 129-35.
Guang-Bin, Huang, Zhu Qin-Yu, and Siew Chee-Kheong. 2004. "Extreme learning machine: a new learning scheme of feedforward neural networks." In 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541), 985-90 vol.2.
Ha, Thien M, and Horst Bunke. 1997. 'Off-line, handwritten numeral recognition by perturbation method', IEEE Transactions on Pattern Analysis and Machine Intelligence, 19: 535-39.
Hagan, Martin T, Howard B Demuth, Mark H Beale, and Orlando De Jesús. 1996. Neural network design (Pws Pub. Boston).
Hinton, Geoffrey E. 2012. 'A practical guide to training restricted Boltzmann machines.' in, Neural networks: Tricks of the trade (Springer).
Hinton, Geoffrey E, and Ruslan R Salakhutdinov. 2006. 'Reducing the dimensionality of data with neural networks', science, 313: 504-07.
Huang, G. B., H. Zhou, X. Ding, and R. Zhang. 2012. 'Extreme learning machine for regression and multiclass classification', IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, 42: 513-29.
Huang, Guang-Bin, Xiaojian Ding, and Hongming Zhou. 2010. 'Optimization method based extreme learning machine for classification', Neurocomputing, 74: 155-63.
Huang, Guang-Bin, Qin-Yu Zhu, and Chee-Kheong Siew. 2006. 'Extreme learning machine: Theory and applications', Neurocomputing, 70: 489-501.
Jiménez Delgado, M. Carmen, and Ignacio Cañas Guerrero. 2007. 'The selection of soils for unstabilised earth building: A normative review', Construction and Building Materials, 21: 237-51.
Kim, Myoung-Jong, Dae-Ki Kang, and Hong Bae Kim. 2015. 'Geometric mean based boosting algorithm with over-sampling to resolve data imbalance problem for bankruptcy prediction', Expert Systems with Applications, 42: 1074-82.
Kotsiantis, Sotiris B, I Zaharakis, and P Pintelas. 2007. 'Supervised machine learning: A review of classification techniques', Emerging artificial intelligence applications in computer engineering, 160: 3-24.
Krosnowski, Adam Davis. 2011. "A proposed best practice method of defining a standard of care for stabilized compressed earthen block production." In Masters Abstracts International. Citeseer.
Manning, Christopher, Prabhakar Raghavan, and Hinrich Schütze. 2010. 'Introduction to information retrieval', Natural Language Engineering, 16: 100-03.
McDonnell, Mark D., Migel D. Tissera, Tony Vladusich, André van Schaik, and Jonathan Tapson. 2015. 'Fast, Simple and Accurate Handwritten Digit Classification by Training Shallow Neural Network Classifiers with the ‘Extreme Learning Machine’ Algorithm', PLoS ONE, 10: e0134254.
Minke, G. 2006. 'Building with Earth Design and Technology of a Sustainable Architecture Birkhäuser–Publishers for Architecture', Berlin, Germany.
Powers, David Martin. 2011. 'Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation'.
Reddy, B. V. Venkatarama, Richardson Lal, and K. S. Nanjunda Rao. 2007. 'Optimum Soil Grading for the Soil-Cement Blocks', Journal of Materials in Civil Engineering, 19: 139-48.
Rigassi, Vincent. 1985. 'Compressed earth blocks: Manual of production', Gesellschaft fur Tecchnische Zusammenaebeit GTZ, Deutsche.
Roux, Nicolas Le, and Yoshua Bengio. 2008. 'Representational Power of Restricted Boltzmann Machines and Deep Belief Networks', Neural Computation, 20: 1631-49.
Salakhutdinov, Ruslan, Andriy Mnih, and Geoffrey Hinton. 2007. "Restricted Boltzmann machines for collaborative filtering." In Proceedings of the 24th international conference on Machine learning, 791-98. Corvalis, Oregon, USA: ACM.
Santoso, B, H Wijayanto, KA Notodiputro, and B Sartono. 2017. "Synthetic over sampling methods for handling class imbalanced problems: A review." In IOP Conference Series: Earth and Environmental Science, 012031. IOP Publishing.
Sitton, Jase D., and Brett A. Story. 2016. 'Estimating Soil Classification Via Quantitative and Qualitative Field Testing for Use in Constructing Compressed Earth Blocks', Procedia Engineering, 145: 860-67.
Sitton, Jase D., Yasha Zeinali, and Brett A. Story. 2017. 'Rapid soil classification using artificial neural networks for use in constructing compressed earth blocks', Construction and Building Materials, 138: 214-21.
Sokolova, Marina, and Guy Lapalme. 2009. 'A systematic analysis of performance measures for classification tasks', Information Processing & Management, 45: 427-37.
Sun, Yanmin, Andrew KC Wong, and Mohamed S Kamel. 2009. 'Classification of imbalanced data: A review', International Journal of Pattern Recognition and Artificial Intelligence, 23: 687-719.
Tissera, M. D., and M. D. McDonnell. 2016a. "Enhancing deep extreme learning machines by error backpropagation." In 2016 International Joint Conference on Neural Networks (IJCNN), 735-39.
Tissera, Migel D., and Mark D. McDonnell. 2016b. 'Deep extreme learning machines: supervised autoencoding architecture for classification', Neurocomputing, 174: 42-49.
Wang, Di, Ping Wang, and Yan Ji. 2015. 'An oscillation bound of the generalization performance of extreme learning machine and corresponding analysis', Neurocomputing, 151: 883-90.
Wang, Yuguang, Feilong Cao, and Yubo Yuan. 2011. 'A study on effectiveness of extreme learning machine', Neurocomputing, 74: 2483-90.
Zarinabad, Niloufar, Martin Wilson, Simrandip K Gill, Karen A Manias, Nigel P Davies, and Andrew C Peet. 2017. 'Multiclass imbalance learning: Improving classification of pediatric brain tumors from magnetic resonance spectroscopy', Magnetic Resonance in Medicine, 77: 2114-24.