Algorithmic prediction in policing: assumptions, evaluation, and accountability
Bibliographic record
Abstract
The goal of predictive policing is to forecast where and when crimes will take place in the future. The idea has captured the imagination of law enforcement agencies around the world. Many agencies are purchasing software tools with the goal of reducing crime by mapping the likely locations of future crime to guide the deployment of police resources. Yet the claims and promises of predictive policing have not been subject to critical examination. This paper provides a review of the theories, techniques, and assumptions embedded in various predictive tools and highlights three key issues about the use of algorithmic prediction. Assumptions: The algorithms used to gain predictive insights build on assumptions about accuracy, continuity, the irrelevance of omitted variables, and the primary importance of particular information (such as location) over others. In making decisions based on these algorithms, police are also directed towards particular kinds of decisions and responses to the exclusion of others. Evaluation: Media coverage of these technologies implies that they are successful in reducing crime. However, these claims are not necessarily based on independent, peer reviewed evaluations. While some evaluations have been conducted, additional rigorous and independent evaluations are needed to understand more fully the effect of predictive policing programmes. Accountability: The use of predictive software can undermine the ability for individual officers or law enforcement agencies to give an account of their decisions in important ways. The paper explores how this accountability gap might be reduced.
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 | 0.268 | 0.600 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.003 | 0.029 |
| Scholarly communication | 0.015 | 0.026 |
| Open science | 0.005 | 0.008 |
| Research integrity | 0.007 | 0.008 |
| Insufficient payload (model declined to judge) | 0.003 | 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".