Preserving the Healthcare Data by Using PPDM Technique

Introduction

In the growing era of technological world, the process of storing information and accessing that information increased rapidly. The information normally falls under the category of business related as well as transactional data (Gupta & Chandra, 2020). Those collected information might be containing sensitive information as well as private data which leads to the privacy issues. It is the technique of uncovering hidden patterns and trends in data. Those information will be gathered from databases and warehouses (Gheware et al., 2014). In this technological era, various business organizations rely on different data mining tools. It involves the process of collecting both structured and unstructured data. The process available will be suitable to process structured data but tedious for unstructured data (Kiranmai & Damodaram, 2014).

Ghate and Ingle (2015) suggested that by applying generalisation and suppression processes to the dataset, the Single Pass K Means Anonymization Algorithm can get an anonymised representation of the data and Canbay et al. (2020) demonstrated a new technique called a Mondrian based utility optimization model. This model will be utilized as a trustworthy tool for outlier detection of the given dataset to promote privacy preserving of those data.

Wang et al. (2020) introduced enhanced Mondrian anonymization model based on Self-Organizing Map (SOM Mondrian). This method aims on reducing the time complexity of Anonymization method and it also helps to increase the accuracy of data mining results. Ouyang, Huang, and Xin (2008) illustrated randomization approach for privacy-preserving mining for sequential patterns. This enhanced to identify the true frequent sequences by modifying every frequent sequence of the dataset. Ong and Miri (2007) have defined a strategy to ensure privacy during testing phase of the database. So, a different approach has been introduced by utilizing a privacy filter to make decision on the results gained by the queries examined. This has been implemented with query modification and data randomization techniques.

Gokulnath et al. (2015), explained about preservation of privacy by utilizing PCA (Principal Component Analysis) based perturbation technique. This method has been employed by K-means cluster algorithm and machine learning based clustering methodology. This results in improvising privacy and performance of data.

Kaur (2017) have exhibited a new method on hybrid approach based on perturbation and suppression technique. This paper enhanced the privacy by overcoming the limitation of information loss by hiding the information of users. Charu, Aggarwal, and Philip (2004) introduced new specific approach for PPDM which does not require problem – solving algorithm because it maps the original dataset into anonymized dataset. This paper illustrates that condensation technique preserves the inter-attribute correlations of the dataset.

Borhade and Shinde (2014), proposed a new approach by utilizing Association Rule with Condensation Approach. This helps to maximize privacy and minimize the data loss by sharing the information but without disclosing the individual’s identity. Patel et al., (2012), introduced a new approach to extract association rule for horizontally partitioned data. This approach is highly useful for protecting our data from intruders and also it provides authentication within the parties involved. Rahman et al. (2010) proposed a new approach called “A Real-Time Privacy-Sensitive Data Hiding Approach Based on Chaos Cryptography” which is used to protect the privacy sensitive data. This approach is based on multiple levels of data hiding models which is highly reliable for protecting the sensitive data of users.