The Methods Of Rule Extraction And Attribute Reduction Based On The Hesitant Fuzzy Linguistic Formal Decision Context

In the big data era with the rapid growth of information,how to extract useful information from various types of data has become one of the hot issues.As an effective mathematical tool for data mining,concept lattice has been applied to many fields.In real life,people always think vaguely,and people often hesitate between several possible linguistic values when evaluating things.To deal with such data,under the hesitant fuzzy linguistic formal context,the hesitant fuzzy linguistic concept lattice is constructed.And we further study the methods of rule extraction and attribute reduction based on the hesitant fuzzy linguistic formal decision context.The main research contents are as follows:1.In order to handle hesitant fuzzy linguistic information better,combining the hesitant fuzzy linguistic term set with the concept lattice theory,the hesitant fuzzy linguistic formal context is studied,and the corresponding concept lattice is constructed.Aiming at the problem that different people will use the HFLTS on different granularity linguistic term set when evaluating things,the transformation method of HFLTS is defined.This method can reduce the information loss,and it is reversible.By changing the threshold value,we can realize the dynamic transformation of the hesitant fuzzy linguistic formal context with different granularity in different situations.2.Based on the hesitant fuzzy linguistic formal context,this thesis further proposes the rule extraction method based on hesitant fuzzy linguistic formal decision context.Firstly,under the hesitant fuzzy linguistic formal decision context,a rule set with uncertainty is obtained based on the hesitant fuzzy linguistic decision concept.Secondly,combining the weighted similarity between the HFLTSs,the rule extraction algorithm for obtaining uncertain rules is proposed by the confidence and support degree of hesitant fuzzy linguistic decision rules.Finally,the rationality and effectiveness of the proposed method are illustrated by an example of online teaching evaluation of teachers.3.To reduce the complexity of concept lattice construction,the attribute reduction algorithm under hesitant fuzzy linguistic formal decision context is studied.Firstly,the reduction algorithms under hesitant fuzzy linguistic formal context are proposed by the score function and deviation function.Secondly,the attribute reduction by means of conditional entropy is conducted in the consistency hesitant fuzzy linguistic formal decision context.Finally,the attribute reduction by means of limitary conditional entropy is studied in the inconsistency hesitant fuzzy linguistic formal decision context.