Privacy Preserving Pattern Based Anonymity (k, P) Anonymity in Time Series Data

摘要

Publishing data from a micro data table containing sensitive attributes, maintaining an individual privacy of that attributes is a difficult task. The k-anonymity model was proposed for privacy preserving data publication. Most of the existing anonymization techniques are used slicing concept in k-anonymity. Thus the Systematic approach used for anonymizing data improves the data utility as well as the quality of the privacy-utility tradeoff. However, in case of existing anonymization methods does not considered pattern based privacy protection of time series data for reduce the information loss and increase the privacy. To overcome this, the proposed system introduces a new anonymization model called pattern based (k, P)-anonymity which is used in time series data. A time series is a set of data regularly gathered at usual intervals for analysis. According to the sensitivity of the diseases the time series micro data table is partitioned into four categories. The proposed pattern based (k, P)-anonymity has two phase. On the first phase, k-anonymity is required for time series in the entire database. On the second phase, P-anonymity is required for the pattern representations associated with each record in a same group. This model publishes both the attribute values and the patterns of time series in separate data forms. Our experimental result shows that the proposed pattern based (k, P)-anonymity data preserve the pattern information for each time series.