Abstract
We have many kinds of data of time series such as daily temperatures and daily stock prices. We have proposed a method to express their linguistic global trend via representative values on the fuzzy intervals in the temporal axis and local features via the position of large difference between the original data and the data generated from the global trend. In the previous research, however, it is applicable only to time series of exact values but not to those of fuzzy values like a stage evaluation. In this paper, we propose a method for linguistic expressions for time series data with fuzziness in the value by applying an extension principle of fuzzy sets to the previous method, where a result with this method is obtained in a fuzzy set of linguistic expressions. The membership values of fuzzy set are approximated by the primary words.
