Learning Product Rankings Robust to Fake Users
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Analytics in the Face of Fraudulent Data This article presents a novel online learning algorithm for identifying optimal product rankings in the presence of fake users and corrupted data. In recent years, e-commerce platforms, such as Amazon, have witnessed a growing number of fake users and click farms. These fraudulent actors seek to boost the position of certain products in the display ordering (i.e., product ranking). Further, platforms’ reliance on data analytics exacerbates the effect of these fake users as machine learning algorithms leverage user feedback to determine product rankings. In the face of these challenges, the present article departs from the status quo that is based on detecting fake users and instead proposes a robust learning methodology. More specifically, the article presents a robust online learning algorithm that converges to the optimal product ranking even when it is impossible to distinguish between real and fake users in the data.
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.
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 | 0.019 | 0.020 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.008 |
| Science and technology studies | 0.005 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.000 | 0.002 |
| Insufficient payload (model declined to judge) | 0.010 | 0.003 |
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 it