Impact of Additive noise on Information loss characteristics of datasets

The prominence of the “big data” notion has dramatically contributed to the growing interest in data mining in recent years. Many different disciplines employ data mining, including business intelligence, scientific discovery, healthcare sectors, and many others as an application-driven discipline. Extraction of knowledge from a large data set is the goal of data mining activities. Although data mining can produce important information for several purposes, people are increasingly growing worried about data mining’s potential hazards to data privacy. To address such issues, privacy-preserving data mining (PPDM) has grown in popularity. The main goal of PPDM is to create data-mining methods that may be used on databases without invading people’s privacy. In this paper, original data was modified by applying additive noise on datasets and its impact on information characteristics are studied. The extent of dispersion or spread in the attribute values caused by noise was also examined. The mean variation error resulted from additive noise shifting the actual value. The median was more robust to outliers and noise, with less impact in additive noise. The mode values were affected minimally, causing data distribution shifts. The value of standard deviation increased.