Graph model building for red tide data based on DWFCM clustering algorithm

Red tide is one of the most important ocean disasters in the world, which has a serious impact on human production and ecological environment. In order to reduce the impact of red tide, it is necessary to build a reasonable and effective storage model for red tide history data so that the forecasters can find out the required data for comparison, forecast the red tide stage, take measures to prevent and cure red tide quickly and accurately. Most people use relational database to store red tide data. This method has a stable table structure but ignores red tide own characteristics when it occurs, and the speed and precision of query red tide need to be improved. In this paper, we research the storage model of red tide data and proposed a method of graph model building for red tide based on DWFCM clustering algorithm. The FCM clustering algorithm was improved by clustering center selection, weighted Euclidean distance and objective function optimization. The improved DWFCM clustering algorithm was used to cluster the red tide data according to the stage, and the clustered data was stored in the graph model. Experiments show that the DWFCM algorithm can cluster red tide data accurately. The graph model builds relationships between each piece of data and each stage, increasing the speed and accuracy of queries and improving forecaster's ability of red tide forecast quickly and accurately.