文献类型：期刊文献

中文题名：基于机器学习理论的海洋水质评价模型

英文题名：Ocean water quality evaluation model based on machine learning theory

作者：张莹[1];谢仕义[1];邓伟彬[1];彭发定[1];余昱昕[2];张培珍[2]

机构：[1]广东海洋大学数学与计算机学院,湛江524088;[2]广东海洋大学电子与信息工程学院,湛江524088

年份：2019

卷号：41

期号：6

起止页码：819

中文期刊名：物探化探计算技术

外文期刊名：Computing Techniques For Geophysical and Geochemical Exploration

收录：CSTPCD、、Scopus

基金：国家自然科学基金(11974084);广东省普通高校特色创新类项目(2018KTSCX091);广东省大学生创新创业训练计划项目(CXXL2019082)

语种：中文

中文关键词：机器学习;海洋水质评价;大样本;支持向量机

外文关键词：machine learning;marine water quality assessment;large sample;support vector machine

中文摘要：为构建适用于海洋水质的评价模型,这里以大数据下的机器学习理论为基础,利用一个包含40万个站位组成,每个站位包含13个水质指标信息的理论假设大样本,选择三种适用于多特征多分类问题的机器学习算法:决策树、贝叶斯和支持向量机来构建适用于水质评价的模型,得出基于支持向量机算法的评价模型效果最好,可作为解决利用多水质指标信息综合评价海洋水质这一问题的有效方法。

外文摘要：An evaluation model is construct suitable for marine water quality based on the machine learning theory under big data in this paper. It is composed of 400,000 stations, each station contains 13 water quality theoretical hypothesis large samples. Three machine learning algorithms including the decision tree algorithm. Bayes and support vector machines which suitable for multi-feature and multi-classification problems are selected. The evaluation model based on the support vector machine algorithm has the best effect. It can be used as an effective method to solve the problem of comprehensive evaluation of marine water quality by using multi-water quality indicators.