Application and Design of Web Information Extraction System Based on Pattern Discovery
Xia Cai · Basic Automation · 2003
It gives the rapid growth of public information sources on the World Wide Web, it is increasingly attractive to extract data from these sources. Current W eb sites present information on various topics in various formats. A great amoun t of effort is often required for a user to manually locate and extract useful d ata from the Web sites.A reference architecture based on pattern d iscovery is developed,which applies PAT trees to pattern discovery.The process r eq uires no human intervention and training example. Experimental result shows that it can achieve high extraction rate over popular search engines.