van Zyl, TL2015-12-182015-12-182014-07Van Zyl, T,L. 2014. A comparison of machine learning techniques for predicting downstream acid mine drainage. In: 35th Canadian Symposium on Remote Sensing (IGARSS) 2014, Quebec, Canada, 13-18 July 2014ftp://ftp.legos.obs-mip.fr/pub/tmp3m/IGARSS2014/abstracts/3311.pdfhttp://hdl.handle.net/10204/832635th Canadian Symposium on Remote Sensing (IGARSS) 2014, Quebec, Canada, 13-18 July 2014. . Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, please consult the publisher's websiteWe consider the challenge of providing downstream predictions of water quality using a time-series of upstream insitu measurements and a time-series of remote sensed precipitation data from the Tropical Rainfall Measuring Mission (TRMM). We use a windowing approach over historical values to generate a prediction for the current value. We evaluate a number of Machine Learning techniques as regressors including Support Vector Regression, Random Forests, Stochastic Gradient Decent Regression, Linear Regression, Ridge Regression and Gaussian Processes. We show that overall we are able to attain R(sup2) values above 0:80 ( 0:16) for most target variables and that Random Forests are the most effective Machine Learning technique in this predictive task.enTropical Rainfall Measuring MissionTRMMMachine learning techniquesWater qualityAcid mine drainageConference Presentationvan Zyl, T. (2014). A comparison of machine learning techniques for predicting downstream acid mine drainage. IGARRS. http://hdl.handle.net/10204/8326van Zyl, TL. "A comparison of machine learning techniques for predicting downstream acid mine drainage." (2014): http://hdl.handle.net/10204/8326van Zyl T, A comparison of machine learning techniques for predicting downstream acid mine drainage; IGARRS; 2014. http://hdl.handle.net/10204/8326 .TY - Conference Presentation AU - van Zyl, TL AB - We consider the challenge of providing downstream predictions of water quality using a time-series of upstream insitu measurements and a time-series of remote sensed precipitation data from the Tropical Rainfall Measuring Mission (TRMM). We use a windowing approach over historical values to generate a prediction for the current value. We evaluate a number of Machine Learning techniques as regressors including Support Vector Regression, Random Forests, Stochastic Gradient Decent Regression, Linear Regression, Ridge Regression and Gaussian Processes. We show that overall we are able to attain R(sup2) values above 0:80 ( 0:16) for most target variables and that Random Forests are the most effective Machine Learning technique in this predictive task. DA - 2014-07 DB - ResearchSpace DP - CSIR KW - Tropical Rainfall Measuring Mission KW - TRMM KW - Machine learning techniques KW - Water quality KW - Acid mine drainage LK - https://researchspace.csir.co.za PY - 2014 T1 - A comparison of machine learning techniques for predicting downstream acid mine drainage TI - A comparison of machine learning techniques for predicting downstream acid mine drainage UR - http://hdl.handle.net/10204/8326 ER -