Verbal Expression of Time Series with Global Trend and Local Features

We have many kinds of time series such as stock prices. We understand them via their verbal expressions in a natural language rather than conventional stochastic models. We propose a method to have a verbal expression for a global trend and local features of time-series data. A global trend is extracted via representative values, e.g. weighted averages, on the fuzzy intervals in the temporal axis and local features are specifled as the positions of large difierences between the original data and the data representing the global trend. We apply the method to the data of Multimodal Summarization for Trend Information (MuST).