1. Title, Journal and Authors
Title: Harmful algal blooms prediction with machine learning models in Tolo Harbour
Authors: Xiu Li, Jin Yu, Zhuo Jia and Jingdong Song
(Shenzhen Key Laboratory of Information Science and Technology, Tsinghua University)
muBP and LMBP, one of the ANN models are introduced. SVM and GRNN are also introduced to compare the algal blooms prediction accuracy.
Water quality data are collected monthly and biweekly, while the meteorological data are daily data. For selecting the more suitable prediction interval, 7-days and 14-days predictions are performed.
In conclusion, (1) Compared to muBP, LMBP shows better performance, (2) GRNN is the fastest, but test results are unsatisfactory, (3) SVM shows the best performance, but consumes the biggest time, and (4) the 7-days prediction is better than the 14-days.
3. Originality & Creativity
- Various machine learning models are introduced to predict optimal results.
- SVM is better predicting model but too expensive to generalize.
Dae Seong Jeong / Integrated Ph.D. program
Environmental Systems Engineering Lab.
School of Environmental Science & Engineering
Gwangju Institute of Science and Technology
1 Oryong-dong Buk-gu Gwangju, 500-712, Korea
Phone : +82-10-2003-7860
E-mail : email@example.com