A paper that I wrote as the principal author along with Dr. Saeed Shaeri and Dr. STMLD Senevirathna from Charles Sturt University has been published online by Springer International. It will also be printed in the Journal of Environmental Science and Pollution Research. The context is the examination of a parametric neural network (Group Method of Data Handling (GMDH)), and its limitations for rainfall modelling. I then introduce a series of potential empirical improvements that will enhance parametric GMDH through the process of hybridising with a selection of algorithms including machine learning algorithms, signal processing techniques, and fuzzy methods. Originally, the paper was 49 pages, but the peer reviewers recommended moving much of the heavy mathematics into a set of appendices, which I did. In case anyone is wondering, I am a civil engineer with specialties in both hydraulics and structural engineering. Dr. Shaeri is a civil engineer and coastal engineering specialist, and Dr. Senevirathna is a civil engineer and water quality specialist. The full paper including appendices can be found at the link below. If you would like to read the paper, I would welcome any comments.
Review of the limitations and potential empirical improvements of the parametric group method of data handling for rainfall modelling | SpringerLink
Ron.
Review of the limitations and potential empirical improvements of the parametric group method of data handling for rainfall modelling | SpringerLink
Ron.