Information fusion and numerical characterization of a multi-source information system
作者：
Highlights：
• In order to take fully advantage of evidence theory and integrate multi-granulation structures, we propose the novel definitions of multi-source rough approximations and corresponding multi-granulation rough approximations, probability distribution and basic probability assignment, then construct the connection between rough approximations and evidence theory.
• The results in (1) are extended to multi-source covering information system.
• Two Shannon’s fusion algorithms based on equivalence relations and coverings, involved in the significance degree of condition attributes set with respect to a sample, conditional probability and information entropy, are presented to measure the classification uncertainty degree of a decision class or a decision partition in a multi-source information system, respectively.
• By combining the significance degree and conditional probability, defined in this paper, we design a multi-granulation probabilistic rough set and considered the relationship with Multi-granulation rough set.
摘要
•In order to take fully advantage of evidence theory and integrate multi-granulation structures, we propose the novel definitions of multi-source rough approximations and corresponding multi-granulation rough approximations, probability distribution and basic probability assignment, then construct the connection between rough approximations and evidence theory.•The results in (1) are extended to multi-source covering information system.•Two Shannon’s fusion algorithms based on equivalence relations and coverings, involved in the significance degree of condition attributes set with respect to a sample, conditional probability and information entropy, are presented to measure the classification uncertainty degree of a decision class or a decision partition in a multi-source information system, respectively.•By combining the significance degree and conditional probability, defined in this paper, we design a multi-granulation probabilistic rough set and considered the relationship with Multi-granulation rough set.
论文关键词：Uncertainty measure,Covering,Multi-granulation rough set,Evidence theory,Multi-granulation variable precision rough set,Information fusion
论文评审过程：Received 6 August 2017, Revised 3 January 2018, Accepted 4 January 2018, Available online 5 January 2018, Version of Record 20 February 2018.
论文官网地址：https://doi.org/10.1016/j.knosys.2018.01.008
