Anonymity Assessment System Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anonymization methods fail to maintain acceptable privacy levels as individuals can still be identified through cross-correlation of non-personally identifiable information in datasets, even after removal of personal identification fields.
Innovation Solution
A system and method using supervised learning techniques to assess the anonymity of datasets by training machine learning models on original and anonymized data, evaluating the quality of anonymization, and providing suggestions for further anonymization based on the used algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If personal identification fields are removed from a dataset, then individual privacy is protected to some extent, but individuals can still be identified through cross-correlation of remaining fields
Solution Approach 1:
The system employs machine learning models that are trained on both original and anonymized datasets to evaluate the effectiveness of anonymization. The models provide feedback by generating evaluation scores that indicate the level of anonymization quality, allowing iterative improvement of anonymization techniques until acceptable privacy protection is achieved.
Solution Approach 2:
The patent replaces manual anonymization assessment with automated machine learning-based evaluation. Instead of relying on manual inspection or simple statistical tests, the system uses trained ML models to automatically assess whether anonymization has been effective, substituting mechanical/manual processes with intelligent automated systems.
2Measurement precision
If machine learning models are trained to assess anonymization quality, then evaluation accuracy is improved, but computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-training machine learning models on datasets before they are needed for evaluation. The models are trained in advance on both original and anonymized datasets, so that when actual anonymization assessment is required, the heavy computational work has already been completed, reducing real-time processing complexity.
Solution Approach 2:
The system changes parameters by using different evaluation metrics and model configurations depending on the specific anonymization scenario. The machine learning models can adjust their evaluation parameters based on the type of data and anonymization technique used, optimizing the balance between evaluation accuracy and computational complexity for each case.
Data Source
AI summary
A method, computer system, and a computer program product for assessing anonymity of a dataset is provided. The present invention may include receiving an original dataset and an anonymized dataset. The present invention may also include preparing a testing dataset and a training dataset for a machine learning algorithm based on the received original dataset and anonymized dataset. The present invention may then include training a machine learning model based on the prepared training dataset. The present invention may further include generating an evaluation score based on the trained machine learning model and the prepared testing dataset. The present invention may also include presenting the generated evaluation score to a user.