Estimating the Credibility of Examples in Automatic Document Classification
Bibliographic record
Abstract
Clustering is a key technique within the KDD process, with k-means, and the more general k-medoids, being well-known incremental partition-based clustering algorithms.A fundamental issue within this class of algorithms is to find an initial set of medians (or medoids) that improves the efficiency of the algorithms (e.g., accelerating its convergence to a solution), at the same time that it improves its effectiveness (e.g., finding more meaningful clusters).Thus, in this article we aim at providing a technique that, given a set of elements, quickly finds a very small number of elements as medoid candidates for this set, allowing to improve both the efficiency and effectiveness of existing clustering algorithms.We target the class of k-medoids algorithms in general, and propose a technique that selects a well-positioned subset of central elements to serve as the initial set of medoids for the clustering process.Our technique leads to a substantially smaller amount of distance calculations, thus improving the algorithm's efficiency when compared to existing methods, without sacrificing effectiveness.A salient feature of our proposed technique is that it is not a new k-medoid clustering algorithm per se, rather, it can be used in conjunction with any existing clustering algorithm that is based on the k-medoid paradigm.Experimental results, using both synthetic and real datasets, confirm the efficiency, effectiveness and scalability of the proposed technique.
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 | 0.009 | 0.055 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".