Enhancing Knowledge Graph-based Credit Default Prediction via Double-Filter Negative Sampling
Bibliographic record
Abstract
Knowledge graphs (KGs) are increasingly used in credit default prediction to model complex data relationships. However, the performance of KG embedding (KGE) is often hindered by low-quality negative samples generated by traditional random negative sampling (RNS) strategies. These low-quality samples limit the model’s ability to learn effective features of data relations. To address this, we propose a Double Filter based Similarity Negative Sampling (DNS) strategy. DNS first employs Bisecting K-means to cluster semantically similar entities, then performs a secondary filtering step to remove overly similar entities from each cluster, thus preventing pseudo-negative examples. We integrate DNS with the TransR model to create the TransR-DNS algorithm, which generates high-quality negative samples to optimise the KGE process. Experiments on a public credit dataset demonstrate that our method consistently outperforms baselines. By replacing traditional sampling with DNS, we improve the prediction performance across various models, validating the effectiveness of our strategy.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| 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 teacher head, 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".