A Scalable ($\alpha, k$)-Anonymization Approach using MapReduce for Privacy Preserving Big Data Publishing

Different tools and sources are used to collect big data, which may create privacy issues. k-anonymity, l-diversity, t-closeness etc. privacy preserving data publishing approaches are used data de-identification, but as multiple sources is used to collect the data, chance of re-identification is very high. Anonymization large data is not a trivial task, hence, privacy preserving approaches scalability has become a challenging research area. Researchers explore it by proposing algorithms for scalable anonymization. We further found that in some scenarios efficient anonymization is not enough, timely anonymization is also required. Hence, to incorporate the velocity of data with Scalable k-Anonymization (SKA) approach, we propose a novel approach, Scalable ( α, k)-Anonymization (SAKA). Our proposed approach outperforms in terms of information loss and running time as compared to existing approaches. With best of our knowledge, this is the first proposed scalable anonymization approach for the velocity of data.