A Comparative Survey Of Algorithmic Feed Recommendation System Designs
Bibliographic record
Abstract
Social media platforms are highly reliant on algorithmic feed systems to deliver content to users. Unlike content recommender systems typically studied in academia, recommendation algorithms for social media feeds are multi-stakeholder and designed to maximize usage, rather than relevance or affinity. How feed algorithms are designed and exactly what content is recommended to users has come under increasing scrutiny from the public and lawmakers. Companies have responded to this scrutiny with more transparency around their systems, including their recommendation algorithms. To aid in comparisons of these newly-transparent systems, we perform a survey of social media feed algorithm systems by conducting a qualitative document analysis of primary source documents. Our survey identifies salient design choices that different apps have made, and algorithm traits that result from those design choices. The key areas of our survey are feed content inventory selection, features used for ranking and four key algorithm traits, along with metrics that capture those traits. We also perform a case study of X’s recently open-sourced feed algorithm, with a particular focus on the key characteristics and algorithm traits identified in our larger survey.
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 | 0.018 | 0.062 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.009 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.008 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".