CAJ | 학술논문

根据海洋环境监测数据的空间特性，提出了监测数据的空间离群点挖掘方法。利用监测站位经、纬度确定空间邻域并建立空间索引，应用空间局部离群系数SLOF （Spatial Local Outlier Factor）的度量方法，利用监测参数值确定空间站位的离群程度，找出与其空间邻域中其它站位的监测参数值存在明显差异的空间站位。研究结果表明，该方法能利用多维监测数据快速有效地挖掘空间局部离群点，为进一步检查出监测数据中的噪声，以及发现其它潜在有用的知识提供服务。
근거해양배경감측수거적공간특성，제출료감측수거적공간리군점알굴방법。이용감측참위경、위도학정공간린역병건립공간색인，응용공간국부리군계수SLOF （Spatial Local Outlier Factor）적도량방법，이용감측삼수치학정공간참위적리군정도，조출여기공간린역중기타참위적감측삼수치존재명현차이적공간참위。연구결과표명，해방법능이용다유감측수거쾌속유효지알굴공간국부리군점，위진일보검사출감측수거중적조성，이급발현기타잠재유용적지식제공복무。
According to the spatial characteristics of marine environmental monitoring data, a spatial local outlier mining method for marine environmental monitoring datasets has been developed. This paper employs the measurement method of SLOF ( Spatial Local Outlier Factor) for the measurement of the spatial outlier score of monitoring data, which uses latitudes and longitudes of monitoring sites to determine the spatial neighborhood and establish the spatial index, then uses the monitoring factors to determine the spatial outlier score. The experimental results show that the proposed SLOF algorithm can effectively mine the spatial local outlier based on multidimensional monitoring data. This method can use for spatial local outlier detection for marine environmental monitoring datasets, and make further efforts for detecting the noise data or revealing potential and meaningful knowledge.