Đăng nhập
 
Tìm kiếm nâng cao
 
Tên bài báo
Tác giả
Năm xuất bản
Tóm tắt
Lĩnh vực
Phân loại
Số tạp chí
 

Bản tin định kỳ
Báo cáo thường niên
Tạp chí khoa học ĐHCT
Tạp chí tiếng anh ĐHCT
Tạp chí trong nước
Tạp chí quốc tế
Kỷ yếu HN trong nước
Kỷ yếu HN quốc tế
Book chapter
Chờ xuất bản
Tạp chí quốc tế 2015
Số tạp chí 525(2015) Trang: 598-606
Tạp chí: Journal of Hydrology
Liên kết:

Increasing the accuracy of pedotransfer functions (PTFs), an indirect method for predicting non-readily available soil features such as soil water retention characteristics (SWRC), is of crucial importance for large scale agro-hydrological modeling. Adding significant predictors (i.e., soil structure), and implementing more flexible regression algorithms are among the main strategies of PTFs improvement. The aim of this study was to investigate whether the improved effect of categorical soil structure information on estimating soil-water content at various matric potentials, which has been reported in literature, could be enduringly captured by regression techniques other than the usually applied linear regression. Two data mining techniques, i.e., Support Vector Machines (SVM), and k-Nearest Neighbors (kNN), which have been recently introduced as promising tools for PTF development, were utilized to test if the incorporation of soil structure will improve PTF’s accuracy under a context of rather limited training data. The results show that incorporating descriptive soil structure information, i.e., massive, structured and structureless, as grouping criterion can improve the accuracy of PTFs derived by SVM approach in the range of matric potential of 6 to 33 kPa (average RMSE decreased up to 0.005 m3 m-3 after grouping, depending on matric potentials). The improvement was primarily attributed to the outperformance of SVM-PTFs calibrated on structureless soils. No improvement was obtained with kNN technique, at least not in our study in which the data set became limited in size after grouping. Since there is an impact of regression techniques on the improved effect of incorporating qualitative soil structure information, selecting a proper technique will help to maximize the combined influence of flexible regression algorithms and soil structure information on PTF accuracy.

Các bài báo khác
Số tạp chí 11(1), June-August, 2015(2015) Trang: 47-51
Tạp chí: American International Journal of Research in Science, Technology, Engineering & Mathematics
Số tạp chí 11(2), June-August, 2015(2015) Trang: 99-103
Tạp chí: American International Journal of Research in Science, Technology, Engineering & Mathematics
Số tạp chí Volume 3 Issue 4(2015) Trang: e00764-15
Tạp chí: Genome Announcements
Số tạp chí 2(2015) Trang:
Tạp chí: Journal of Optimization Theory and Applications
Số tạp chí 247(2015) Trang: 774-781
Tạp chí: European Journal of Operational Research
Số tạp chí 2(2015) Trang: 1-12
Tạp chí: Renewables Wind, Water, and Solar
Số tạp chí XVIII(2015) Trang: 147-165
Tạp chí: Transactions on Computational Collective Intelligence
Số tạp chí Vol. 4(1)(2015) Trang: 1-6
Tác giả: Hồ Thanh Thâm
Tạp chí: Nova Journal of Engineering and Applied Sciences
Số tạp chí Volume 15, Issue 4, Version 1.0(2015) Trang: 1-4
Tạp chí: Global Journal of Human Social Sciences
Số tạp chí 20(1)(2015) Trang: 101-122
Tác giả: Võ Văn Dứt
Tạp chí: The Asian Academy of Management Journal
Số tạp chí 10(1), March-May 2015(2015) Trang: 1-6
Tạp chí: American International Journal of Research in Science, Technology, Engineering & Mathematics
Số tạp chí 10(1), March-May 2015(2015) Trang: 73-78
Tạp chí: American International Journal of Research in Science, Technology, Engineering & Mathematics

Crossref DOI of CTUJoS


BC thường niên 2018


Con số ấn tượng (VN | EN)


Bản tin ĐHCT


TCKH tiếng Việt


TCKH tiếng Anh

 
 
Vui lòng chờ...