Building predictors from vertically distributed data
Bibliographic record
Abstract
Due in part to the large volume of data available today, but more importantly to privacy concerns, data are often distributed across institutional, geographical and organizational boundaries rather than being stored in a centralized location. Data can be distributed by separating objects or attributes: in the homogeneous case, sites contain subsets of objects with all attributes, while in the heterogeneous case sites contain subsets of attributes for all objects. Ensemble approaches combine the results obtained from a number of classifiers to obtain a final classification. In this paper, we present a novel ensemble approach, in which data is partitioned by attributes. We show that this method can successfully be applied to a wide range of data and can even produce an increase in classification accuracy compared to a centralized technique. As an ensemble approach, our technique exchanges models or classification results instead of raw data, which makes it suitable for privacy preserving data mining. In addition, both final model size and runtime are typically reduced compared to a centralized model. The proposed technique is evaluated using a decision tree, a variety of datasets, and several voting schemes. This approach is suitable for physically distributed data as well as privacy preserving data mining. Copyright c ○ 2004 Sabine McConnell and David
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 | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.002 |
| 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".