Efficiently answering top-k typicality queries on large databases
Bibliographic record
Abstract
890 Finding typical instances is an effective approach to understand and analyze large data sets. In this paper, we apply the idea of typicality analysis from psychology and cognition science to database query answering, and study the novel problem of answering top-k typicality queries. We model typicality in large data sets systematically. To answer questions like “Who are the top-k most typical NBA players?”, the measure of simple typicality is developed. To answer questions like “Who are the top-k most typical guards distinguishing guards from other players?”, the notion of discriminative typicality is proposed. Computing the exact answer to a top-k typicality query requires quadratic time which is often too costly for online query answering on large databases. We develop a series of approximation methods for various situations. (1) The randomized tournament algorithm has linear complexity though it does not provide a theoretical guarantee on the quality of the answers. (2) The direct local typicality approximation using VP-trees provides an approximation quality guarantee. (3) A VP-tree can be exploited to index a large set of objects. Then, typicality queries can be answered efficiently with quality guarantees by a tournament method based on a Local Typicality Tree data structure. An extensive performance study using two real data sets and a series of synthetic data sets clearly show that top-k typicality queries are meaningful and our methods are practical.
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 | 0.004 | 0.024 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.002 | 0.005 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.009 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.002 | 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".