Ordinal Data Stream Collection with Condensed Local Differential Privacy
Bibliographic record
Abstract
Continuous collection of users' ordinal data is essential to numerous industrial applications, such as disease surveillance, road traffic analysis, and computation offloading. However, the data collector is not fully-trusted, so that the necessitate privacy protection on the sensitive ordinate data streams of users becomes critical. Local Differential Privacy (LDP) is typically used to resolve this problem with high efficiency for real-time data analysis. However, existing LDP methods are mainly designed for one-time data collection, and they bring low utility in some time slots for continuous collection, due to the data sparsity problem that some users' data are empty in most of time slots. Therefore, we propose a Condensed LDP (CLDP)-based scheme, which provides real-time statistics with the protection of each user's time-series data and the high utility in each time slot. First, our scheme optimizes the utility by allocating each user's privacy budget in the time dimension. The privacy budgets originally used in empty time slots are saved and allocated to non-empty time slots for every user, so as to increase the utility in each time slot. Then, CLDP is used to further counteract the negative impact of data sparsity. Finally, privacy analysis is given to show the privacy protection of data streams, and sufficient experiments on real datasets are conducted to demonstrate the effectiveness of the proposed scheme.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.002 | 0.006 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".