A Time Series Data Mining System

Abstract

The invention provides a time series data mining system which is wide in application range, easy to understand, convenient to operate and suitable for small and medium-sized enterprise competitive intelligence analysis systems. The system comprises an initial data processing module, a time series cluster analysis module, a single time series pattern discovery module and a same-category multiple time series pattern discovery module which are sequentially connected, wherein the initial data processing module acquires time series data in the access time sequence and divides and smooths the time series data; the time series cluster analysis module analyzes the similarity of the received time series data, clusters the time series data with a clustering algorithm and divides the time series data into a first-category time series......an (N-1)-category time series and an N-category time series according to correlations; the single time series pattern discovery module arranges a first-element pattern set......an (N-1)-element pattern set and an N-element pattern set; the same-category multiple time series pattern discovery module calculates time series frequent patterns.

Description

technical field [0001] The invention relates to the technical field of data mining, in particular to a time series data mining system. Background technique [0002] Data mining refers to the non-trivial process of revealing implicit, previously unknown and potentially valuable information from a large amount of data in a database. Data mining is a decision support process, which is mainly based on artificial intelligence, machine learning, pattern recognition, statistics, database, visualization technology, etc., to analyze data highly automatically, make inductive reasoning, and dig out potential patterns from it. Help decision makers make the right decisions. [0003] The amount of data in the database is very large, and certain rules must be followed when analyzing the data. Due to the ubiquity of time series, the commonly used data mining is time series data mining, which collects data at different time points. This type of data reflects the state or degree of a certai...

AI Technical Summary

Problems solved by technology

[0004] Time series data mining is the most commonly used means of existing data mining. However, the existing time series data mining methods are highly targeted, and the time series mining methods and models between different environments and different objects often vary greatly. For example, seasonal Rainfall, changes in China's GDP, financial market analysis, etc. all require different time series data mining models

In addition, with the increasingly fierce competition among enterprises, more and more attention has been paid to the analysis of competitive intelligence. However, the existing competitive intelligence data analysis system is very expensive and unaffordable for small and medium-sized enterprises. They urgently need a simple, fast and convenient data analysis system. mining method

Images