Application of Regularized Extreme Learning Machine Based on BIC Criterion and Genetic Algorithm in Iron Ore Price Forecasting
Futian Weng, Muzhou Hou, Tianle Zhang, Yunlei Yang, Zheng Wang, Hongli Sun, Hao Zhu, Jianshu Luo
Available Online July 2018.
- https://doi.org/10.2991/msam-18.2018.45How to use a DOI?
- BIC criterion; genetic algorithm; regularized extreme learning machine; iron ore price forecast
- Forecasting international iron ore is a well-known issue, BIC criterion is used to select the relevant variables of iron ore price. On the basis of the traditional extreme learning machine (ELM), the regular term is introduced to control the complexity of the model, and the genetic algorithm (GA) is used to regularize the extreme learning machine. The input-layer weight matrix and the hidden-layer threshold matrix of the (RE-ELM) model are optimized to establish a BIC-based genetic algorithm and a regularization extreme learning machine (BIC-GA-RELM) iron ore price prediction model to increase the performance of the RE-ELM model. The results show that BIC-GA-RELM model has achieved the state of art performance, then a new method is provided for iron ore price prediction.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Futian Weng AU - Muzhou Hou AU - Tianle Zhang AU - Yunlei Yang AU - Zheng Wang AU - Hongli Sun AU - Hao Zhu AU - Jianshu Luo PY - 2018/07 DA - 2018/07 TI - Application of Regularized Extreme Learning Machine Based on BIC Criterion and Genetic Algorithm in Iron Ore Price Forecasting BT - Proceedings of the 2018 3rd International Conference on Modelling, Simulation and Applied Mathematics (MSAM 2018) PB - Atlantis Press SP - 212 EP - 217 SN - 1951-6851 UR - https://doi.org/10.2991/msam-18.2018.45 DO - https://doi.org/10.2991/msam-18.2018.45 ID - Weng2018/07 ER -