| Hidden Web contains lots of well-structured and high-quality information. Along with the enhancement of IT application, the quantity of such information has been increasing faster and faster. Although the Hidden Web information is increasing rapidly, the quantity indexed by the search engine is very small. So it causes a lot of information waste.This paper first analyses the origin of Hidden Web and search methods mainly on Hidden Web database classification and search interface integration. And then in order to find a common search method for Hidden Web, two major searching technologies are studied:(1) Automatic Hidden Web search interface recognition. Without using sample set, only by submitting keywords and analyzing the results, the method could find the Hidden Web search interface rapidly and accurately.(2) Hidden Web search keywords selection algorithm. First using sample estimate method to find the search keywords, and then analyzing the words'frequency in the sample set to get the formula of the sample frequency. Keywords selected with the formula reflect the trend of their frequency in the Hidden Web database, so the selected words could be the best selection.The method of interface recognition and keywords selection algorithm is tested by certain experiments. The experiments well validate our research. |
PDF Full Text Request