Research On Multi-Strategy Schema Matching Based On Fuzzy Decision

With the rapid development of computer networks and database systems, enterprise competition and increased merger, Web information integration and enterprise information integration needs of more urgent. In the information integration process, due to the surrounding environment of the different data sources may be relational databases, object-oriented database, XML, HTML, etc., this heterogeneity of information resources everywhere, but more and more applications need to access a variety of heterogeneous data sources, heterogeneous data integration process issue has become urgent to solve.Schema matching is the basic way to solve the problem. Machine learning technology, graph theory, information retrieval technology is widely adopted in the study of schema matching. In different domains and from different levels to solve the problem of schema matching the application of these technologies achieved good results. At the same time in solving schema matching problem these technologies also have their own limitations and flaws. In this paper, we use multi-strategy approach to solve the problem of schema matching in the web information integration. In multi-strategy matching system, the multi-strategy is not always better than a single strategy. For a matching task, how to choose the best combination of strategies? In this paper, the sub-strategy of the recommended way to select the most suitable for the current matching task sub-strategy portfolio and fuzzy comprehensive evaluation method to the various sub-strategies similarity judge In brief, we make the following contributions in this paper:1. We propose multi-strategy based on fuzzy decision schema matching model, introduce the composition of the model briefly. This matching model utilizes the model of heterogeneity between the characteristics of a wide range of information on utilization patterns, from the schema element name, schema structure and schema instances of three angles of linguistics-based strategy, structure-based strategy, instances-based The strategy of the two models match the results of similarity between the calculations, effectively improved the accuracy of matching results.2. We use fuzzy decision schema matching similarity ideas on a comprehensive evaluation to improve the schema matching similarity calculation of the degree of automation to improve the matching process efficiency and matching the quality of results. We introduce how to calculate the similarity of the schema using multi-strategy schema matching model.3. In order to detect the optimal set of strategies we have improved a detection algorithm, and it improves the efficiency of the detection.