A recommendation and risk classification system for connecting rough sleepers to essential outreach services
Open accessAbstract
Rough sleeping is a chronic experience faced by some of the most disadvantaged people in modern society. This paper describes work carried out in partnership with Homeless Link (HL), a UK-based charity, in developing a data-driven approach to better connect people sleeping rough on the streets with outreach service providers. HL's platform has grown exponentially in recent years, leading to thousands of alerts per day during extreme weather events; this overwhelms the volunteer-based system they currently rely upon for the processing of alerts. In order to solve this problem, we propose a human-centered machine learning system to augment the volunteers' efforts by prioritizing alerts based on the likelihood of making a successful connection with a rough sleeper. This addresses capacity and resource limitations whilst allowing HL to quickly, effectively, and equitably process all of the alerts that they receive. Initial evaluation using historical data shows that our approach increases the rate at which rough sleepers are found following a referral by at least 15% based on labeled data, implying a greater overall increase when the alerts with unknown outcomes are considered, and suggesting the benefit in a trial taking place over a longer period to assess the models in practice. The discussion and modeling process is done with careful considerations of ethics, transparency, and explainability due to the sensitive nature of the data involved and the vulnerability of the people that are affected.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Metrics at k are more commonly used in ML for public policy to address real-world resource constraints.
“Importantly, we included metrics at a certain percentage of the population (often referred to as metrics at k) because they are more commonly used in ML for public policy in order to specifically addresses real-world resource constraints [2,12,66].”
From It’s Just Not That Simple: An Empirical Study of the Accuracy-Explainability Trade-off in Machine Learning for Public Policy (2022) · cites this paper as [66]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
(2003) · Journal of Machine Learning Research · doi:10.5555/944919.944937 · Cited by 32,337 · see citing works
Software Framework for Topic Modelling with Large Corpora
Rek, R. R. U., & Sojka, P. (2010) · doi:10.13140/2.1.2393.1847 · Cited by 3,766 · see citing works
Cross‐validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure
Roberts, D. R., Bahn, V., Ciuti, S., Boyce, M. S., Elith, J., Guillera‐Arroita, G., Hauenstein, S., Lahoz‐Monfort, J. J., Schröder, B., Thuiller, W., et al. (2017) · Ecography · doi:10.1111/ecog.02881 · Cited by 3,689 · see citing works
Bias in error estimation when using cross-validation for model selection
Varma, S., & Simon, R. (2006) · BMC Bioinformatics · doi:10.1186/1471-2105-7-91 · Cited by 2,485 · cited 1× · see citing works
“Due to the temporal dependencies present in the data, a month forward-chaining temporal cross validation approach [12, 15] was adopted to ensure our model error estimates were robust across the entirety of the dataset.”
MethodAn Improved Non-monotonic Transition System for Dependency Parsing
Honnibal, M., & Johnson, M. (2015) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d15-1162 · Cited by 618 · see citing works
Toward an International Understanding of Homelessness
Toro, P. A. (2007) · Journal of Social Issues · doi:10.1111/j.1540-4560.2007.00519.x · Cited by 211 · see citing works
Homelessness in Europe and the United States: A Comparison of Prevalence and Public Opinion
Toro, P. A., Tompsett, C. J., Lombardo, S., Philippot, P., Nachtergael, H., Galand, B., Schlienz, N., Stammel, N., Yabar, Y., Blume, M., et al. (2007) · Journal of Social Issues · doi:10.1111/j.1540-4560.2007.00521.x · Cited by 136 · see citing works
Meinbresse, M., Brinkley‐Rubinstein, L., Grassette, A., Benson, J. B., Hall, C., Hamilton, R., Malott, M., & Jenkins, D. M. (2014) · Violence and Victims · doi:10.1891/0886-6708.vv-d-12-00069 · Cited by 81 · see citing works
Women and Rough Sleeping : A Critical Review of Current Research and Methodology
Bretherton, J., & Pleace, N. (2018) · White Rose Research Online (University of Leeds, The University of Sheffield, University of York) · Cited by 28 · see citing works
Housing insecurity, homelessness and populism : evidence from the UK
Fetzer, T., Sen, S., & Souza, P. (2019) · Warwick Research Archive Portal (University of Warwick) · doi:10.82444/warw.33797917 · Cited by 22 · see citing works
Design choices for productive, secure, data-intensive research at scale in the cloudPreprint
Arenas, D., Atkins, J., Austin, C., Beavan, D., Cabrejas-Egea, A., Carlysle-Davies, S., Carter, I. M., Clarke, R., Cunningham, J. A., Doel, T., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.08737 · Cited by 7 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 1
- Wilde, H., Chen, L. L., Nguyen, A., Kimpel, Z., Sidgwick, J., De Unanue, A., Veronese D, Mateen B, Ghani R and Vollmer, S. (2020) A Recommendation and Risk Classification System for Connecting Rough Sleepers to Essential Outreach Services. arXiv:2007.15326. · arXiv
Cited by
Sorted by citations
Bell, A., Solano-Kamaiko, I., Nov, O., & Stoyanovich, J. (2022) · 2022 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3531146.3533090 · Cited by 103 · see citing works
The Possibility of Fairness: Revisiting the Impossibility Theorem in Practice
Bell, A., Bynum, L., Drushchak, N., Zakharchenko, T., Rosenblatt, L., & Stoyanovich, J. (2023) · 2023 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3593013.3594007 · Cited by 21 · see citing works
- Authors
- Wilde, H., Chen, L. L., Nguyen, A., Kimpel, Z., Sidgwick, J., De Unanue, A., Veronese, D., Mateen, B., Ghani, R., & Vollmer, S.
- Affiliations
- University of Warwick (GB)
- Northwestern University (US)
- The Alan Turing Institute (GB)
- Harvard University (US)
- Carnegie Mellon University (US)
- Corresponding author
- Vollmer, Sebastian
- Venue
- Data & Policy
- Publisher
- Cambridge University Press (CUP)
- Published
- 2021
- Volume
- 3
- ISSN
- 2632-3249 (ISSN-L)
- OpenAlex
- W3045855879
- Cited by
- 3
- References
- 11
- Citations by year
- 2023: 1 · 2022: 1 · 2021: 1
- FWCI
- 0.85
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Data & Policy (publishedVersion, cc-by)
- Warwick Research Archive Portal (University of Warwick) (acceptedVersion)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Alan Turing Institute
- University of Warwick
- University of Edinburgh
- Topics
- Health Sciences › Health Professions › General Health Professions › Homelessness and Social Issues
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Information Systems › Data Mining and Machine Learning Applications
- Keywords
- rough sleeping, homelessness, outreach services, alert prioritization, risk classification, human-centered machine learning, volunteer capacity, explainable AI, vulnerable populations
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
