Research On Object-Level Keyword Search Algorithm Over Graph Databases

The integration of relational database technology and information retrieval technology has rapid development driven by the application requirements. Users who need neither understand the complex structured query language, nor know the underlying database schema, can query data in the database like using a Web search engine. About the relational database information retrieval strategies, domestic and foreign experts and scholars put forward a number of different points of view. Among them, there are both tuple leveled and object leveled. The amount of data in relational databases growing makes the scale of data graph increasing, information retrieval efficiency is getting lower and lower. The field of relational database information retrieval is facing the challenge of big data has become an unavoidable fact.As the graph database technology continues to be matured, its field of application is constantly expanding. Innate flexible graph model not only meets the application requirements of social networking sites, but also adapts to the graph algorithms strongly. Through studied graph database techniques and full-text indexing techniques, and analyzed the research status of object-level information retrieval and graph database retrieval, an extraction method of the conversion from relational data to graph data is proposed, and an improvement for the existing object-level modeling is made, an object level information retrieval algorithm embedded graph database is designed, compared to the tuple level information retrieval, the object level information retrieval has many advantages such as small-scale data graph, high results integrity, and non-reproducible results, etc.. The algorithm on the basis of the importance of considering the retrieval keyword, and using heuristic rules query. Combined with the graph database and the relational database, the relational database information retrieval can provide an effective solution for mass data conditions, and expand the application areas of the graph database.In order to verify the effectiveness of the algorithm and the availability of the prototype system, DBLP dataset is used to validate the effectiveness and efficiency of the algorithm’s query results by experiment.P@k is used to validate the retrieval efficiency of the algorithm, and a comparison and analysis is made for the algorithm’s retrieval efficiency. The final experimental results show that the graph database object level keyword search algorithm performs effectiveness, and will have a perfect application prospect.