基于数据挖掘的船舶入侵数据建模方法

Data mining-based modeling method for ship invasion data

在船舶入侵数据建模过程中,入侵特征和入侵数据检测模型参数均对船舶入侵数据建模效果产生影响,而当前船舶入侵数据建模方法没有考虑入侵特征和入侵数据检测模型参数之间的联系,这样无法建立有效的船舶入侵数据检测模型。为了克服当前船舶入侵数据建模方法存在的缺陷,以改善船舶入侵数据建模的整体性能,提出基于数据挖掘的船舶入侵数据建模方法。首先分析当前国内外船舶入侵数据建模的研究现状,找到入侵特征和入侵数据检测模型参数之间的联系,然后引入数据挖掘方法对入侵特征和入侵数据检测模型参数之间联系进行建模,找到两者的最优组合,然后根据最优入侵特征和入侵数据检测模型参数建立入侵数据检测模型,并与其他入侵数据建模方法性能进行对比测试。测试结果表明:本文方法获得的入侵数据特征更优,入侵数据检测时间更短,根据最优参数获得更高正确率的入侵数据检测结果,是一种可信度高的入侵数据建模技术。

In the process of ship intrusion data modeling, both intrusion characteristics and intrusion data detection model parameters have an impact on the effect of ship intrusion data modeling. However, the current method of ship intrusion data modeling does not consider the relationship between intrusion characteristics and intrusion data detection model parameters, so it is impossible to establish an effective ship intrusion data detection model, in order to overcome the current ship intrusion. In order to improve the overall performance of ship intrusion data modeling, a data mining-based method for ship intrusion data modeling is proposed. Firstly, the research status of ship intrusion data modeling at home and abroad is analyzed, and the relationship between intrusion characteristics and intrusion data detection model parameters is found. Then,data mining method is introduced to model the relationship between intrusion characteristics and intrusion data detection model parameters, and the optimal combination of the two is found. Then, the intrusion is established according to the optimal intrusion characteristics and intrusion data detection model parameters. The data detection model is tested and compared with other intrusion data modeling methods. The intrusion data feature obtained by this method is better, the intrusion data detection time is shorter, and the intrusion data detection results with higher accuracy can be obtained according to the optimal parameters. This method is a highly reliable intrusion data modeling technology.