RED TIDE PREDICTION METHOD USING NEURAL NETWORK AND SVM

摘要

PURPOSE: red tide occurrence prediction method, using neural network and support vector machines (damage) with minimize by red tide forecast red tide occur based on marine environment. ;CONSTITUTION: learning stuff prediction red tide generation machined (100) in advance. Red tide generating state prediction processing input data, by referring to learning stuff (200). Red tide predicts to include that penalty term and the calm distance that can be changed which presentation are mistakenly positioned from focal plane to substance using SVM models when occurring. ;The 2013 of copyright KIPO submissions;[Reference numerals] (100) pre-process; (200) nerve network/SM; (AA) learning stuff; (BB) learn; (CC) input material; (DD) generation of red tide is predicted