forked from ChelseaKR/nearmiss
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathCITATION.cff
More file actions
68 lines (68 loc) · 3.1 KB
/
Copy pathCITATION.cff
File metadata and controls
68 lines (68 loc) · 3.1 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
cff-version: 1.2.0
message: >-
If you use nearmiss — the software, the pipeline, or the published open
dataset it produces — please cite it using the metadata below. nearmiss is
both a software project and a dataset: the analysis code and the
aggregated, minimum-occupancy open dataset (GeoJSON plus data card) are
released together. When citing a specific published dataset release, also cite the
versioned GeoJSON and its accompanying data card (docs/DATA-CARD.md), and
state the exposure source and its date alongside any rate you quote, since
every rate here is normalized by an exposure estimate and carries a
confidence interval.
title: nearmiss
abstract: >-
nearmiss is an open dataset and a statistically honest analysis of road
hazards and near misses for safe-streets advocacy. It turns the close-pass,
door-zone, blind-corner, and surface-hazard reports that cyclists and
pedestrians already make into a rigorous, open, reusable dataset and an
analysis that refuses to lie with a heat map. Reports are validated against a
published schema, deduplicated, geocoded, snapped to street segments,
classified, and quality-flagged through a recorded, reproducible pipeline.
Counts are normalized by an exposure estimate so that risk is reported as a
rate, never as raw report volume; every published rate, ranking, and
comparison carries a confidence interval and an n; reporting bias is named
rather than hidden; and statistically significant clusters are identified
with Getis-Ord Gi* rather than inferred from kernel-density appearance alone.
Contributor privacy is protected: precise raw reports stay private, and the
open dataset is aggregated to public street segments with low-count segments
withheld (k-anonymity); no precise coordinate or timestamp is published.
type: software
authors:
- given-names: Chelsea
family-names: Kelly-Reif
alias: ChelseaKR
version: 0.4.0
date-released: '2026-08-16'
license: Apache-2.0
repository-code: https://github.com/ChelseaKR/nearmiss
url: https://nearmiss.chelseakr.com
keywords:
- road safety
- near miss
- cycling
- pedestrian safety
- open data
- spatial statistics
- Getis-Ord
- exposure normalization
- advocacy
preferred-citation:
type: dataset
title: 'nearmiss: open near-miss and road-hazard dataset'
abstract: >-
The published nearmiss dataset: aggregated, minimum-occupancy open data
(GeoJSON) plus its data card (docs/DATA-CARD.md), produced by the nearmiss pipeline
from validated, deduplicated, geocoded, and segment-snapped hazard reports.
Counts are normalized by an exposure estimate and reported as rates with a
confidence interval and an n; cite the versioned release and state the
exposure source and its date alongside any rate you quote.
authors:
- given-names: Chelsea
family-names: Kelly-Reif
alias: ChelseaKR
version: 0.1.0
date-released: '2026-06-16'
license: Apache-2.0
url: https://nearmiss.chelseakr.com/fars/national/
repository: https://github.com/ChelseaKR/nearmiss
# TODO(#184): assign a DOI (e.g. via Zenodo) against a shipped tag and set `doi:` here.