Affordable Access

Characteristics of Java class file for code optimization in a distributed computing environment

Murdoch University
Publication Date
  • Computer Science


al. on: rtainty Assessment using Neural Networks and Interval sutrosophic Sets for Multiclass Classification Problems KRALPEERAPUN CHUN CHE FUNG KOK WAI WONG School of Information Technology Murdoch University Perth, Western Australia AUSTRALIA {p.luaipeerapun I 1.fung I [email protected] paper presents an approach to multiclass classification. A pair of k-class neural networks are t k pairs of truth membership and false membership values. The k pairs of errors in the prediction ut patterns are also estimated using interpolation techniques. Two techniques are proposed for the fication in this paper. First, the truth and false memberships are compared in order to classify the le classes. Second, estimated errors are used to weight the degrees of truth and false mem- ults of the combination between weighted truth and weighted false memberships are ification. The estimated errors as well as the difference between the truth and false membership dered as the elements of the indeterminacy membership used to identify level of uncertainty in the ification. Together the three membership values form interval neutrosophic sets. We experiment to the classical benchmark problems including balance, ecoli, glass, lenses, wine, yeast, and zoo machine learning repository, Our approach improves classification performance compared to an que which applied only to the truth membership created fiom a single k-class neural network. :- multiclass classification, uncertainty, interval neutrosophic sets, multiclass neural network, feedfor- pagation neural network always exist in the real world data and ing required to make predictions. Un- oceurs when there is a lack of information world for deciding if the statement is true of uncertainty in the t task. Degree of uncer- f quality in the classification. to represent uncertainty in the classifica- lass neural network classification involves a1 networks that map the input feature network output contain

There are no comments yet on this publication. Be the first to share your thoughts.