Authors
Belousov R.
|
Degree
|
PhD in Technique, Civil Defence Academy EMERCOM of Russia |
|---|---|
|
E-mail
|
romabel-87@mail.ru |
|
Location
|
Khimki |
|
Articles
|
Construction of fuzzy linguistic variables using cluster data analysisThe paper describes an approach for constructing fuzzy linguistic variables, based on the use of
cluster analysis. The basic assumption in this method is the fact that the distance from the center of
the cluster element degree of membership fuzzy set is reduced. The maximum value of membership
will have a core (average) of cluster. Degree of membership of the other elements decreases
linearly with distance from the cluster core. The initial set of data is divided into clusters. Each cluster
corresponds to a linguistic (verbal) characteristic, which corresponds to a fuzzy linguistic variable.
The normalized distance from the cluster members to its center corresponds to the degree accessory
to the term: the greater the distance, the lower the degree of membership. The analytical form
of the membership functions is obtained by fitting the results of cluster analysis. This approach allows
you to automate the process of developing management systems with fuzzy logic.
Read more...
|
