Schema extracting and query processing for semistructured data in COMMIX

Information integration over heterogeneous data sources in the Internet environment is a major concern of the database community. One of the most significant problems concerning this subject is schema extracting and query processing for semistructured data. COMMIX, is a massive information integration system, which is developed by the Research Group of Content-Oriented Massive Information Integration, Analyzing, Processing and Services. The paper describes novel methods on integration-oriented local schema extracting and pipelining-based query processing. It also illustrates the work of the algorithm in COMMIX with examples. According to the experimental data, it proves that the efficiency of local schema extracting algorithm is higher than traditional global schema extracting algorithm in semistructured database. Finally it proves the correctness of pipelining-based query processing algorithm and communication complexity and computation complexity.EI0SUPPL.230-2361