Tên bài báo:
A Novel Hybrid Model of Rotation Forest Based Functional Trees for Landslide Susceptibility Mapping: A Case Study at Kon Tum Province, Vietnam- Tác giả:
- Phạm Thái Bình
- Tham gia cùng:
- Ngô Thị Thanh Hương
- Tạp chí:
- International conference on Geo-spatial technologies and Earth resources
- Năm xuất bản:
- 2017
- Trang:
- Từ trang 186 đến trang 201
- Lĩnh vực:
- Kỹ thuật xây dựng công trình giao thông
- Phạm vi:
- Quốc tế
Tóm tắt:
In this study, we proposed a novel hybrid model namely Rotation Forest based Functional Trees (RFFT), which is a hybrid intelligent approach of two state of the art machine learning techniques of Functional Trees (FT) classifier and Rotation Forest (RF) ensemble, for landslide susceptibility mapping at the Kon Tum Province, Viet Nam. Landslide affecting factors (slope angle, slope aspect, elevation, valley depth, land use, NDVI, soil type, lithology, distance to geology boundaries, and distance to faults), and 1404 past and current landslide locations have been first collected from the study area for generating training and testing datasets. Secondly, the hybrid model RFFT has been constructed for landslide susceptibility assessment using training dataset. Performance of the proposed RFFT model has been validated by analysis of the Receiver Operating Characteristic (ROC) curve and statistical indexes, and compared with a well-known landslide models namely Support Vector Machines (SVM) and the single FT. Results show that the proposed RFFT model has good performance for landslide susceptibility assessment. It has better predictive capability compared with well-known SVM model and single FT model. Therefore, it can be concluded that the proposed RFFT model should be used as a great alternative method for better landslide susceptibility assessment in landslide prone area.
