Currently, the information is available in the organizations to gain more utilities together with some reason such as improving the strategy of organization, research, or etc.,. For this reason, the data holder cannot avoid to share some existing data-collecting of organization to be outside the scope. Hence, before the data is shared to be outside the scope of data-collecting, the data holder must be confidence the shared data cannot be re-identified to the owner again. The (k,e)-Anonymous is one of the most of prominent models in term of privacy preservation. However, the traditional (k,e)-Anonymous model is designed to support only the numeric-sensitive attribute. In the real-world, the domain of sensitive attribute can be other data domains such as the categorical-sensitive, ordinal-sensitive, or etc.,. Therefore, in this work, we propose the extended capability of (k,e)-Anonymous to supports the sensitive attribute that is the categorical data-type in conjunction with the DFS algorithm and the post-order traversal.

Currently, the data can be gathered, analyzed, and utilized easier than ever with the aiding of Big Data technologies such as mobile devices, elastic computing platform, or convenient software tools. Thus, privacy of such data could become a bigger issue as well. In this paper, we propose to extend the capability of a prominent privacy preservation model, (k, e)-anonymous to further provide a better option for privacy preservation. We propose to add a support for the privacy-sensitive ordinal data-type to such model, since it originally supports only numerical data. The experiments are conducted to show the characteristics of the modified model. From the results, we can conclude that the characteristics after our work has been applied are very similar to the original, and thus it can be effectively applied to the privacy problem.