The entire planet is online and connected and we are also collectively generating immeasurable amounts of data. As this data is stored online and segregated for easy retrieval but with data exploitation complexity of privacy and other threat also increase. This article stressed the importance of the data-deidentification model.
The Key Takeaways from Article are Here-
- Data deidentification is the process of separating the personal identity of an individual from their data. And with the current status of Machine Learning(ML) technologies, it is easy to detect patterns and identify people based on personal information provided. Hence it is important to put regulations on these models.
- Now with data-deidentification models, to reduce certain information going here and there. HIPAA recommends two approved methods for data de-identification. These methods are expert determination and safe harbor methods.
- Companies can either choose to remove data or their identifiers completely from their records or they can use de-identification API to remove these identifiers from their datasets. However, the first method is effective but you might want to retrieve the data in-house for diverse searches at that time the second option might get difficult.
Read the full article here: