论文部分内容阅读
为了解决在语义W eb信息处理中本体的频繁存取造成的性能问题,提出基于B ayes ian决策理论的本体缓存模型。该模型有效利用本体的语义性和本体的存取日志来抽取特征向量(包括语义特征、存取特征和类型特征),通过B ayes ian决策理论指导在本地缓存频繁使用的本体,并通过机器学习优化缓存模型,提高本体概念和实例缓存命中率。本体的有效缓存减少了本体网络访问的开销,实验表明,采用该本体缓存模型后,原型系统的本体访问速度在G auss分布的本体访问概率下提高了25%左右。
In order to solve the problem of ontology frequent access in semantic W eb information processing, an ontology caching model based on Bayesian decision theory is proposed. The model effectively extracts the feature vectors (including semantic features, access features and type features) by using the ontology semantics and ontology’s access log, and caches the frequently used ontologies locally by B ayes ian decision theory and learns through machine learning Optimize the cache model to improve ontology concepts and instance cache hit ratio. The effective caching of ontology reduces the overhead of ontology network access. Experiments show that using ontology caching model, the ontology access speed of prototype system increases about 25% under the ontology access probability of G auss distribution.