基于贝叶斯算法的森林成熟预测研究

The Forest Maturity Forecast Based on Bayesian Algorithm

我国的森林资源调查和监测工作起步较早,具有丰富的森林资源监测基础数据资料,激增的数据背后隐藏着许多重要的信息,简单的查询和统计已经无法满足林业的需求,需要出现一种挖掘数据背后隐藏的知识手段.该文将数据挖掘的方法引入林业应用,从而为林业的经营决策提供一条新的思路.通过挖掘林龄与其它调查因子间的关系来建立贝叶斯算法预测模型,利用其它调查因子较准确、有效率地预测林龄;确定了与林龄相关性最高的因子:胸径、树高、公顷畜积和郁闭度,解决了传统测量林龄的困难,并可进一步确定森林是否成熟,为合理采伐提供依据.

Data mining method was applied in the forestry,thus a new way of thinking was provided in the decision-making for forestry operators. A bayesian algorithm prediction model was created by mining relationships between age and other factors. It can forcast forest age more accurately and efficiently by the use of the other factors. It determines that the most relevant factors of the age are DBH,tree height,livestock acre plot and canopy density. It can solve the problems of measuring age and determine whether the forest is mature or not. It provides a basis way for a reasonable harvest.