SYSTEM AND METHOD FOR RECOMMENDING PUBLIC DATA GENERALIZATION LEVEL GUARANTEEING DATA ANONYMITY AND USEFUL TO DATA ANALYSIS

摘要

The present invention relates to a system and a method for generating public data useful for data analysis while ensuring data anonymity. The system comprises a generalization level defining unit which defines a K-value as a K-anonymity condition for original data and a generalization level acceptable range for each quasi-identifier attribute; a generalization level combination generating unit which generates all possible generalization level combinations within the defined generalization level acceptable range for each attribute; an anonymity rate and balance rate calculating unit for each generalization level combination which calculates an anonymity rate and a balance rate for each of the generated generalization level combinations for each attribute; and a generalization level combination recommending unit which evaluates a value of each combination in consideration of the calculated balance rate and anonymity rate for each combination, and provides a predetermined number of generalization level combinations in order of value. Accordingly, it is possible to have enough information that data should have in order to use public data again for data analysis while preventing identification of sensitive information of the data.