Feature extraction method suitable for time series data in field of cultural science and technology fusion

Abstract

The invention discloses a feature extraction method suitable for time series data in the field of cultural science and technology fusion, and the method comprises the following steps: 1, obtaining the time series data from a target database, classifying a sequence according to a data type, and obtaining text data and numerical data; 2, classifying the numerical data according to time granularity to obtain macroscopic time series data and microcosmic time series data; the method comprises the following steps: firstly, obtaining time sequence data from a target database, classifying sequences according to data types, obtaining text data, firstly obtaining a high-frequency vocabulary set by using a bag-of-words model, and then carrying out secondary filtering on the vocabulary set by using improved TF-IDF to obtain hot vocabularies; if new data exists, executing the step 1-4 again by using a sliding window; if no new data exists, stopping; the method can quickly process and analyze the time series data, and is helpful for making strategic decisions of enterprises.

Description

technical field [0001] The invention relates to a feature extraction method suitable for time series data in the field of cultural and technological integration. Background technique [0002] With the progress of our country's science and technology and the adjustment of economic development strategy, the development of emerging industries and enterprises with culture as the core and technology as the means has officially become a major situation advocated by various places. On the one hand, general enterprises mainly use financial data or user behavior modeling when predicting their own development hotspots, and do not pay special attention to multi-angle mining of time series data; on the other hand, general time series data mining Because it does not have the characteristics of cultural and technological fusion data, it does not perform data mining on the classification of time series data from time granularity. For an industry or enterprise, the amount of information ca...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a feature extraction method suitable for time series data in the field of cultural and technological integration, which solves the problem of time series data with different time granularities having different important information and less sample data in the prior art; common time series feature mining The algorithm only mines time series information from a single perspective, and cannot fully obtain the value of the data