Deep Web Database Classification Based on Tuple-Matching

Provided by: Binary Information Press
Topic: Data Management
Format: PDF
DeepWeb database classification method based on tuple-matching is proposed in this paper, which selects attributes, attribute instances, relations between attributes, relations between attributes and attribute instances to describe query pattern of query interface, uses depth and frequency of attribute and attribute instance to calculate corresponding weight, construct triple set and tetrad set of query interface based on label text, weight and path of attribute and attribute instance, calculates the correlation of query interfaces by matching triples and tetrads, at last classifies query interface by the correlation.

Find By Topic