A Layered Intersection Scan Algorithm Based on Semantic Context

Along with the rapid growth of the xml data quantity on the Internet, the xml data retrieval research has attracted more and more attention. The searching algorithm based on key words is a research hotspot in this field. We present a context-based layered intersection scan algorithm (CLISA), which uses the context semantic of key words to filter large amount of redundant information, different from the approach of using key words to get SLCA. Based on LISA, CLISA operates LCAs to get SLCAs of different semantics, and then chooses the SLCA that best fits user´s need. Through experiments we can see that CLISA can be more in accordance with the user´s demand on condition that it expends more time than LISA.