Analysis of The Impact of Deep Learning-Based Recommendation Algorithms on Demographic Groups
Bibliographic record
Abstract
This study compares the performance of TensorFlow Recommender (TFRS), Light Factorization Machine (LightFM), and Weighted Matrix Factorization (WMF) on the MovieLens 25M dataset. It focuses on accuracy and fairness across different user groups. Experiments show that TFRS achieves good accuracy and keeps fairness across gender and age, but its performance drops sharply in sparse environments. LightFM performs better in cold-start cases but shows large gaps in fairness, especially among older users. WMF shows the most consistent fairness across age and gender groups because it uses confidence-weighted feedback methods, though its accuracy is lower. In controlled tests, TFRS ranks first in recommendation accuracy, WMF ranks first in exposure balance, and LightFM ranks first in new user adaptability. These results show that each model has strengths depending on the deployment environment. TFRS is good for mobile apps with quick user updates, WMF suits systems with high fairness needs, and LightFM is good when handling new users.
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 | 0.008 | 0.025 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| 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 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".