基于本体的组合预测预警系统模型

Combination Forecasting and Early-Warning System Model Based on Ontology

在分析现有预测预警系统不足的基础上,提出一种基于本体的组合预测预警模型(OCFWSM),解决组合预测预警系统内各单一预测模型和指标体系间的语义异构问题,通过建立本体知识库,实现系统内数据、模型、知识的一致化表示,实现各模型和指标体系间的知识共享和重用。具体分析了本体知识库的组成及其实现原理和方法,根据本体建模理论,建立了领域指标本体、指标本体、预测任务本体及案例本体等本体模型,并利用本体编辑工具Protege3.3.1将本体形式化,选用Pellet作为内部本体推理机,合理选择单一预测模型和指标体系来进行组合预测预警,从而提高预测预警的准确性。最后给出了一个原型系统,具有十分广阔的应用前景。

After analyzing the insufficiency in the present system , a model of on ontology basis is proposed, this model can solve thesemantic heterogeneity of each kind of sole forecasting model and indicator-system in combined forecasting and early-warning system, through the establishment of ontology knowledge base, make possible the consistent representation of data, model and knowledge and the knowledge sharing and reusing. The implementation of principles and methods of ontology knowledge base is analyzed, according to the methodology about ontology modeling, some ontology models are modeled, and the ontology is formalized by Protege3.3.1, using Pellet as an internal ontology inference engine, choosing single forecasting model and indicator-system for combined forecasting and early-warning reasonably, improve the accuracy of forecasting and early-warning, meanwhile, A prototype system？ is also presented and the applying prospect will be well.