基于SPA和QoX的不一致性消除算法

Inconsistency Elimination Algorithm Based on SPA and QoX

在情景感知计算(Context Aware Computing,CAC)的研究与应用过程中,不可避免地会遇到情景信息的不确定性问题。而情景信息不一致性是不确定性问题中最难解决的突出问题,因此如何合理、有效地处理情景信息的不一致性尤为重要。针对情景信息不一致性问题,提出了一种基于集对分析(Set Pair Analysis,SPA)和情景信息综合质量指标(Quality of Comprehensive IndeXes,QoX)的不一致性消除算法,其使用QoX对情景信息综合质量进行度量和评估,提高了情景信息综合质量度量的完整性和精确性;同时,该算法使用SPA对所求质量指标从同一度、对立度、差异度3个方面进行定量分析,从而保证情景信息质量计算过程所考虑的因素更为全面。实验结果表明,相比传统的不一致性处理算法,所提算法的判决正确率有明显提升。

In the research and application of context-aware computing(CAC),it is inevitable to face the problems of uncertainty for context information.Context inconsistency is the most difficult problem to solve in the uncertainty problems.How to deal with context inconsistency reasonably and effectively is an important problem that must be solved.In this paper,a new inconsistency elimination algorithm based on set pair analysis(SPA)and quality of comprehensive indexes(QoX)is proposed to solve the inconsistency problem.The algorithm uses some parameters of QoX to evaluate the overall quality of contexts collected by the context sources,which improves the completeness and accuracy of the evaluation of overall quality of context.In addition,the algorithm uses SPA to quantitatively analyze the quality index from the similarity,opposition and difference,and takes into account the comprehensiveness in the calculation process of quality of context.Experimental results show that the proposed algorithm has higher accuracy than traditional processing algorithms.