基于PSO-BP算法的地理本体概念语义相似度度量

发布时间：2019-06-18 13:05

【摘要】：针对现有度量方法中考虑因素不够全面和因子权重计算依据经验确定的不足,提出粒子群优化BP神经网络(PSO-BP)的地理本体概念语义相似度度量模型。该模型利用本体属性、本体结构和语义关系的相似度,结合权重信息计算概念的综合相似度;同时,利用粒子群算法优化的BP神经网络获取因子权重,避免现有方法中因子权重确定的人为主观干扰。最后,从基础地理信息概念中提取出200组样本,用其中190组作为训练集,对神经网络模型进行训练,以获取权重;剩余10组作为测试集。将该模型和几种常用算法进行对比,通过分析测试集的各算法求解结果和专家判定结果之间的相关系数,结果表明该模型计算地理本体概念的相似度更为准确,符合人类认知特性,效果更好。
[Abstract]:In view of the fact that the factors in the existing measurement methods are not comprehensive enough and the calculation of factor weights is determined by experience, a semantic similarity measurement model of geographical ontology concept based on particle swarm optimization BP neural network (PSO-BP) is proposed. The model uses the similarity of ontology attribute, ontology structure and semantic relationship, combined with the weight information to calculate the comprehensive similarity of the concept. At the same time, the BP neural network optimized by particle swarm optimization algorithm is used to obtain the factor weight to avoid the artificial subjective interference determined by the factor weight in the existing methods. Finally, 200 groups of samples are extracted from the concept of basic geographic information, 190 of which are used as training sets to train the neural network model to obtain the weight, and the remaining 10 groups are used as the test set. The model is compared with several common algorithms, and the correlation coefficients between the results of each algorithm and the results of expert decision are analyzed. The results show that the model is more accurate in calculating the similarity of geographical ontology concepts, in line with the cognitive characteristics of human beings, and the effect is better.
【作者单位】：信息工程大学;海军出版社;
【分类号】：TP391.1;TP18

【参考文献】