AI for Predictive Privacy Risk Assessment

Muhammad Naveed Sajjad, Muhammad Irshad and Muhammad Faisal Shafiq
Additional contact information
Muhammad Naveed Sajjad: Yanal Finance Company, Riyadh, Saudi Arabia
Muhammad Irshad: RMG Company, Riyadh, Saudi Arabia
Muhammad Faisal Shafiq: Lucid Motors, Riyadh, Saudi Arabia

European Journal of Information Technologies and Computer Science, 2026, vol. 6, issue 2, 9-16

Abstract: Organizations that handle sensitive data require a privacy risk assessment, which is very important, but the available traditional approaches cannot be considered predictive. This study was conducted on the application of artificial intelligence (AI) in predictive privacy risk assessment data based on the world’s largest data breaches and hacks dataset. The data were cleaned and standardized to minimize the magnitude of breaches, create numeric attributes of text entries, and create variables such as the organization’s history of breaches, industry, and sensitivity of data. Quantiles of lost log-transformed records were used to categorize breaches into Low, Medium and High-risk. Logistic Regression and Random Forest models were used as supervised learning models and trained to predict risk levels. Logistic Regression offered a linear baseline with reasonable performance, whereas random forest modeled non-linear relationships that enhanced the classification of breaches, especially medium-risk breaches. The analysis of feature importance found that the strongest predictors of privacy risk are data sensitivity, history of organizational breaches, industry, and hacking methods. The findings reveal that AI-driven models are capable of materializing breach classifications and offering understandable information to mitigate risks in time. The research shows that a predictive privacy risk assessment framework can be replicated and has the potential to assist in organizational cybersecurity governance and decision-making using AI.

Keywords: Data breaches; data sensitivity; random forest; risk assessment (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://eu-opensci.org/index.php/compute/article/view/70294 Abstract page (text/html)
https://eu-opensci.org/index.php/compute/article/download/70294/14453 Full text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:epw:comput:v:6:y:2026:i:2:id:70294

DOI: 10.24018/compute.2026.6.2.70294

Access Statistics for this article

More articles in European Journal of Information Technologies and Computer Science from European Open Science
Bibliographic data for series maintained by Support Team ().