Educational Information In A Relational Database Full-text Search Efficiency Improvements And Implementation

The research of full-text search have become hot in recent years, and made many achievements and breakthroughs. In the field of education, a large amount of educational information is stored in database, the two types of information are traditional structured information and the text of unstructured information. For getting the information which we need from database in the short time, is a problem which we have to deal with, and is the real significance of this paper.In recent years, after continuous research and improvement, the text database overall performance has made remarkable progress and it can cope with most demands of the full-text search. However, it is too hard to meet requirement of the joint inquiries of full text search and structure data retrieval for many text databases which has single function. At present, only some large commercial relational database (Oracle,DB2,etc) for full-text search and the query of structure data have good support. But these databases are expensive that most people can't affordable. Today, the open source of relational database with MySQL and PostgreSQL can support full-text search and meet requirement of the joint inquiries of full-text search and structure data. However, when we have a vast amount of information and need to order the output,the performance of full-text search often can't meet the demand. Thus, improving the efficiency of search has very important practical significance.This paper will be in the case of PostgreSQL. Firstly, this paper introduces the PostgreSQL and the search flow of PostgreSQL FTS(full-text search),analysis the structure of index GIN and performance. This paper brings forward a program which is based on inverted index to improve the performance of PostgreSQL full-text search. This program includes the structure of inverted index transformation, scan method and so on. Finally, with testing by large amounts of data we find the efficiency of full text search of improved index structure has been significantly improved and meet the need of educators, prove the feasibility and value of this paper.