An Efficient Hybrid Privacy Preservation Model for Big Data Analytics

Personal information is an important aspect of today’s life. Big data privacy is crucial where structured and unstructured information is frequently produced, processed, and stored in large amounts. Data is increasing in various fields. As the amount of big data grows, there are multiple challenges, and preserving big data privacy is important in big data analytics. This paper summarizes various anonymization techniques for keeping big data privacy which helps to eliminate privacy risks in data preparation. Proposes a new efficient Hybrid Privacy Preservation Model (HPPM) that combines features of k-anonymity and t-closeness. The algorithm has been evaluated by applying various data privacy preservation metrics. The experimental analysis and results show that due to anonymization of data, the information loss is increased but, it preserves data privacy and accuracy of the model at a high level. The HPPM is a better anonymization technique to safeguard the quasi-identifiers and sensitive attributes that reveal user identity.