Dunbo Cai
ABSTRACT
Personal information protection is becoming so important for individuals. Besides personal identifier information (PII), quasi-identifier information (QII) also needs protection, as the community argues, and the solution methods have attracted many research. Many existing methods for protecting QII focus on structured text data which is organized by tables of records. However, free text data that contains QII, is very common in application domains, such as data lakes of a company. The protection of QII in free text data thus need new methods. Supervised machine learning based solutions are promising while usually require a large scale dataset to train the model. Here we propose a novel method towards building such a desired dataset. Our method exploits an existing structured text dataset, a table to sentence generation deep learning model, and incorporated the idea of Piecewise Convolution Neural Network (PCNN). The resulted dataset contains more than 120,000 free text sentences, and many of them contains QII data.
- El Emam, Khaled, and Fida Kamal Dankar. 2008. Protecting privacy using k-anonymity. Journal of the American Medical Informatics Association, 15, 5, 627-637.Google Scholar
- Sweeney, Latanya. 2002. k-anonymity: A model for protecting privacy. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 10, 05, 557-570.Google Scholar
- Šarčević, Tanja, David Molnar, and Rudolf Mayer. 2020. An Analysis of Different Notions of Effectiveness in k-Anonymity. In Proceedings of International Conference on Privacy in Statistical Databases. Springer, Cham, 121-135.Google Scholar
- Machanavajjhala, Ashwin, Daniel Kifer, Johannes Gehrke, and Muthuramakrishnan Venkitasubramaniam. 2007. l-diversity: Privacy beyond k-anonymity. ACM Transactions on Knowledge Discovery from Data (TKDD), 1, 1, 3-es.Google Scholar
- Nininahazwe, Franck Seigneur. 2019. Studying L-Diversity and K-Anonymity Over Datasets with Sensitive Fields. In Proceedings of the International Conference on Artificial Intelligence and Security. Springer, Cham, 63-73.Google Scholar
- Neamatullah, Ishna, Margaret M. Douglass, H. Lehman Li-wei, Andrew Reisner, Mauricio Villarroel, William J. Long, Peter Szolovits, George B. Moody, Roger G. Mark, and Gari D. 2008. Automated de-identification of free-text medical records. BMC medical informatics and decision making, 8, 1, 1-17.Google Scholar
- Iwendi, Celestine, Syed Atif Moqurrab, Adeel Anjum, Sangeen Khan, Senthilkumar Mohan, and Gautam Srivastava. 2020. N-sanitization: A semantic privacy-preserving framework for unstructured medical datasets. Computer Communications, 161, 160-171.Google ScholarCross Ref
- Liu, Zengjian, Buzhou Tang, Xiaolong Wang, and Qingcai Chen. 2017. De-identification of clinical notes via recurrent neural network and conditional random field. Journal of biomedical informatics, 75, S34-S42.Google ScholarDigital Library
- Yogarajan, Vithya, Bernhard Pfahringer, and Michael Mayo. 2020. A review of automatic end-to-end de-identification: Is high accuracy the only metric?. Applied Artificial Intelligence, 34, 3, 251-269.Google ScholarCross Ref
- Zeng, Daojian, Kang Liu, Yubo Chen, and Jun Zhao. 2015. Distant supervision for relation extraction via piecewise convolutional neural networks. In Proceedings of the 2015 conference on empirical methods in natural language processing. Association for Computational Linguistics, Lisbon, Portugal, 1753-1762.Google Scholar
- Liu, Tianyu, Kexiang Wang, Lei Sha, Baobao Chang, and Zhifang Sui. 2018. Table-to-text generation by structure-aware seq2seq learning. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18).AAAI Press, New Orleans, Louisiana, USA, 4881-4888.Google Scholar
- Puduppully, Ratish, Li Dong, and Mirella Lapata. 2019. Data-to-text generation with content selection and planning. In Proceedings of the Thirty-Third AAAI conference on artificial intelligence (AAAI-19).AAAI Press, Hawai, USA, 33, 01, 6908-6915.Google Scholar
- [Online]. Available: http://archive.ics.uci.edu/ml/datasets.php.Google Scholar
Recommendations
Anonymization of Sensitive Quasi-Identifiers for l-Diversity and t-Closeness
A number of studies on privacy-preserving data mining have been proposed. Most of them assume that they can separate quasi-identifiers (QIDs) from sensitive attributes. For instance, they assume that address, job, and age are QIDs but are not sensitive ...
Anonymization of Quasi-Sensitive Attribute Sets in Aggregated Dataset
The widespread use of Internet of Things (IoT) and Data Fusion technologies make privacy protection an urgent problem to be solved. The aggregated datasets generated in these two scenarios face extra privacy disclosure. We define attribute sets with ...
Privacy Preserving Publishing on Multiple Quasi-identifiers
ICDE '09: Proceedings of the 2009 IEEE International Conference on Data EngineeringIn some applications of privacy preserving data publishing, a practical demand is to publish a data set on multiple quasi-identifiers for multiple users simultaneously, which poses several challenges. Can we generate one anonymized version of the data ...
Comments