High-Dimensional Similarity Searches Using A Metric Pseudo-Grid
Bibliographic record
Abstract
Despite the proposal of numerous tree-based access structures for high dimensional similarity searches, techniques based on a sequential scan have been shown to be simple yet quite efficient alternatives. Given that random accesses to disk are expensive, a linear scan of the (smaller) pre-processed dataset is often much more efficient than even a relatively small number of random disk accesses yielded by tree-based indices. In this paper we present a technique which uses a pseudo-partition of a general metric space analog to the VA-file’s partition of the vector space. The rationale is to use a number of pivot objects in the metric space, each one determining a number of hyper-rings in this space. The intersection of those rings, determine pseudo-cells analog to the VA-file cells in the vector space. In order to speedup query processing the data set is clustered (using any applicable clustering technique). Clusters not intersecting cells intersected by the query region cannot contribute to the answer set. Thus, only a few clusters are searched using an I/O efficient linear scan of the cluster’s data. The proposed technique, which we call the M-GRID, is, by construction, applicable to both general metric spaces and to traditional vector spaces as long as a metric distance function is used. The M-GRID is robust to several parameters and experiments with synthetic and real data sets show that it is able to perform nearest neighbor queries up to 10 times faster than the VA-File.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.000 |
| 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".