At the same time,optimal reduction of decision table based on granularity entropy is used to acquire the fuzzy diagnosis rules of ethylene cracking process.

This paper,from the information theory,the concept of information granularity,granularity function and granularity entropy are defined.

This paper discusses the correlation between the number of cut points,granular entropy and classification accuracy in discretization.

With the gradual increase or decrease of attribute in a attribute set leading to different partitions of the universe,this paper proposes a new metric,granular entropy,to measure the capacity of the attribute set for discerning objects when its attribute increases or decreases,and provides a reduction algorithm of information system based on the granular entropy.

The paper offer a new heuristic attribute reduction algorithm based on conditional granularity entropy,thou.

Based on sample entropy-a modified algorithm of approximate entropy,we introduce the defination, algorithm and application of multiscale entropy(MSE),which is applicable to the analysis for the complexity of physiologic signals.

Firstly,multiscale entropy characteristics of several typical nonlinear series were studied,and then based on this,the fluctuant conductance signals of 144 two-phase flow conditions were analyzed,which were collected by using array conductance sensors in upward vertical gas-liquid two-phase flow.

The algorithm steps of multiscale entropy are given.

Representation of time-frequency by using frequency hopping signals through entropy measure and SQP