基于PageRank的地理信息服务质量数据缺失处理

Treatment of Missing Data in Geographic Information Services Quality Based on PageRank

针对现有的大量地理信息服务存在质量要素数据缺失,且直接影响了地理信息服务质量评价的有效性问题,提出了一种基于PageRank的地理信息服务质量数据缺失处理方法。借鉴PageRank算法思想,结合两者之间的相似之处,在计算质量要素缺失数据时,同时考虑要素数据缺失的服务质量以及与其相似的服务的数量和质量,从而提高质量数据的完整度和准确度,便于后续地理信息服务质量的评价。选取德国人工智能研究中心发布的基于OWL-S的地理信息服务语义匹配测试集OWLS-TC4中有关地理信息领域的服务作为测试用例,通过与平均值填充法在数据缺失处理偏差值和准确率的对比分析实验,进一步验证了本文提出的基于PageRank的地理信息服务质量数据缺失处理方法的有效性。

Aiming to the quality factor data missing existed in a large number of geographic information services, and the completeness of the data’s quality factor directly affects the effectiveness of the evaluation of geographic information service quality. A missing data processing method to the quality of geographic information service based on PageRank is proposed in this paper. When calculating the data missing quality element, this paper took into account the quality of service missing factor data and the quantity and quality of similar services, to improve the integrity and accuracy of quality data and facilitate the subsequent evaluation of the quality of geographic information services. This paper selected the geographic information services as a test case issued by the German Center for Artificial Intelligence Research Center, which were geographic information service semantic satching test set based on the OWL-S, by comparing the mean value filling method with the deviation of the data loss and the accuracy of the experimental analysis, validating the effectiveness of the proposed method.