Research on Quality Control for Attribute Data Based on Stratified Sampling Theory

At present, quality contral for attribute data of GIS is a hot topic in the field of geographic information system; and thereinto, quality contral theory for attribute data based on disfigurement rate develops rapidly and it has been used in some fields. In this article, on the basis of disfigurement rate theory of attribute data by stratified random sampling, a disfigurement classification for attribute data from college database is proposed; the disfigurement rate and its variance in all layers and types are provided; the quality condition of attribute data in the whole system is evaluated by disfigurement tolerance.