Abstract
ABSTRACT Attribute reduction for set-valued data commonly took into account the distance or similarity between attribute values. However, little attention has been paid to the problem that sample labels can affect attribute reduction. This paper studies the attribute reduction for set-valued data based on prediction label. Firstly, the prediction label of samples in a set-valued decision information system (SVDIS) is defined. And then, the tolerance relation in an SVDIS based on prediction labels is given, which can distinguish samples not only by the distance between the attribute values, but also by the prediction labels. As a result, some related concepts have been redefined. Moreover, attribute reduction algorithms in an SVDIS based on dependence and decision error rate are designed. Eventually, experimental analysis on real data sets indicates that the designed algorithms can effectively reduce the number of attributes, and improve the classification accuracy in most cases.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
