Marine information management system based on time series outlier detection technology

Abstract

The invention provides a marine information management system based on the time series outlier detection technology. The system mainly comprises six functional modules, i.e. a data acquisition module, a data preprocessing module, an outlier detection module, a data memory module, a data management module and a data display module. The quality of marine data directly affects the scientificalness of the marine information management system. Because the data volume of a marine information database is huge, the marine information database is often susceptible to noise, data loss and data inconsistency, and therefore the invention provides a cumulative variation-based time series outlier detection method; the cumulative variation-based time series outlier detection method can effectively detect outliers in marine data, an appropriate correction method is then chosen to correct the outliers in order to provide clean, accurate and simple data for the marine information database, and thereby the complete set of marine information management system is established to effectively manage marine data, thus providing powerful support for the construction of digital ocean in China.

Description

technical field [0001] The invention relates to a preprocessing method of marine information data, a time series anomaly detection technology and a management method of marine data. Background technique [0002] At present, the research of data mining mainly focuses on the discussion of data mining algorithms, but neglects the research of data preprocessing. However, the data in the actual system generally seldom directly meet the requirements of the data mining algorithm, seriously affecting the execution efficiency of the data mining algorithm, and even causing the deviation of the mining results. According to statistics, the time and cost of data preprocessing account for about 60% of the whole process of data mining. Therefore, effective induction and preprocessing of data sources has become a key issue in the process of data mining system implementation. [0003] With the country's increasing demand for sea use and the rapid development of the marine economy, higher r...

AI Technical Summary

Problems solved by technology

The quality of marine data directly affects the scientific decision-making of the marine information management system. At present, there is no systematic marine data quality control method in China. Manual correction is generally used for control. For a large amount of marine data, the time based on the cumulative change is used. Sequence anomaly detection technology, which detects abnormalities in the collected marine data, stores qualified data and abnormally corrected legal data in the marine information database, manages and displays the marine data

Images