A new data filling approach based on probability analysis in incomplete soft sets. (1st December 2021)

• Record Type: Journal Article
• Authors: Kong, Zhi; Zhao, Jie; Wang, Lifu; Zhang, Junjie
• Is Part Of: Expert systems with applications. Volume 184(2021)
• Publication Date: 2021-12-01
• Subjects: Incomplete soft set -- Data filling -- Strongest association degree -- Probability -- Accuracy; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2021.115358
• Languages: English
• ISSNs: 0957-4174

Abstract

Abstract: Incomplete information is a common phenomenon in practical situations, and even happens in the uncertain problems. Soft set theory shows good performance in dealing with uncertain problems. While the problem of incomplete information appears in soft set. The existing approach can only handle with incomplete data in some special cases with low accuracy. For these deficiencies, we propose a new approach. The proposed method can solve some cases which can't be solved by the existing approach and avoid the influence of the threshold with subjective factors. Further error rate of prediction results can be reduced to the minimum by using the proposed approach. To verify the effectiveness and feasibility of new method, we compare proposed method with existing approach. Experimental results based on UCI benchmark database are shown to demonstrate the performance of new approach.