In heterogeneous data integration, an effective machine learning model plays an important role in schema mapping. Schema mapping machine learning model and its probability learning improvement are analyzed in this paper firstly, and then the concept of multi -analyzer model with the method of fuzzy comprehensive evaluation is put forward to improve machine learning results’ efficiency and accuracy. Comprehensive experiments on multi-analyzer model confirm the effectiveness of multi -analyzer model.

The main challenges that marine heterogeneous data integration faces are the problem of accurate schema mapping between heterogeneous data sources. In order to improve the schema mapping efficiency and get more accurate learning results, this paper proposes a heterogeneous data schema mapping method basing on multianalyzer machine learning model. The multianalyzer analysis the learning results comprehensively, and a fuzzy comprehensive evaluation system is introduced for output results’ evaluation and multi factor quantitative judging. Finally, the data mapping comparison experiment on the East China Sea observing data confirms the effectiveness of the model and shows multianalyzer’s obvious improvement of mapping error rate.

This paper presents a XML system framework of the marine metadata for application optimization of marine metadata database and its standardizing definitions. And gives a detailed XML schema description in detail for the framework. Finally a 3-layer process structure model based on the XML metadata management system is put forward to improve the efficiency of storing and managing the marine metadata.

The existing database systems are designed based on the physical characteristics of the magnetic disks, and they have failed to obviously improve the performance of solid-state disks (SSD). Therefore, a new connection algorithm, namely CSub-Join, is proposed in this paper. It is a query optimization algorithm for flash databases based on a column storage model. The algorithm first reads columns from two joining tables as CSub tables, and then executes joint operations on CSub tables to generate a join index table and a fetch sequence table. Finally, according to the two generated tables, the query results are taken back from the two joining tables by columns. Using the small size of column oriented storage and the high-speed random read of SSD, the algorithm overcomes the speed mismatch between database systems and flash memories. The results of experiments show that CSub-Join algorithm clearly outperforms CSub-Join algorithm under various selectivities and memory sizes.