Fuzzy Prediction Model in Privacy Protection: Takagi–Sugeno Rules Model via Differential Privacy
Bibliographic record
Abstract
Rule–based fuzzy models have modular architectures and come with well-developed design methodologies such that they can build accurate models with good interpretabilities in system modeling. However, a large amount of private data needs to be used for statistical analysis and forecasting in rule–based fuzzy models. The purpose of this study is to build an intelligent model with high accuracy and versatility under the premise of data privacy and model security. To mitigate the risk of malicious attacks on privacy data during the analysis process, we have employed highly regarded differential privacy techniques to devise a novel rule-based fuzzy modeling approach. We propose a function approximation mechanism to reconstruct the objective function and add a perturbation mechanism to the objective function in Takagi-Sugeno rules model. Taking into account the delicate balance between data privacy and utility, we have innovatively introduced a Takagi-Sugeno rule-based model based on differential privacy. This model is applicable to both linear and nonlinear systems, offering protection to sensitive data privacy and model security within the system. We investigate the relationship between the interpretability of the model and the degree of privacy protection. By constructing a reasonable rule base, we achieve higher accuracy than other system modeling methods based on differential privacy. This paper compares the influence of the number of rules on differential privacy, and considers the algorithm performance under various noise distributions. It is shown that the Takagi-Sugeno rules model based on differential privacy has a strong ability to predict and analyze data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.010 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".