Enhancing Epidemic Forecasting: Evaluating the Role of Mobility Data and Graph Convolutional Networks

Furao Shen; Jian Zhao; Suhan Guo; Zhenghao Xu

REVIEW 4 major objections 5 minor 47 references

Enhancing Epidemic Forecasting: Evaluating the Role of Mobility Data and Graph Convolutional Networks

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Mobility data and GCN layers fail to lift COVID-19 forecast accuracy

desk verdict A useful empirical null result about mobility data and GCNs in epidemic forecasting, but the abstract overstates the negative claim: the paper shows 'no evidence' rather than 'no effect'. read the letter →

arxiv 2506.11028 v1 pith:3ZVPQRWD submitted 2025-05-20 cs.LG cs.AI

Suhan Guo , Zhenghao Xu , Furao Shen , Jian Zhao This is my paper · ORCID

classification cs.LGcs.AI

keywords COVID-19forecastingmobilitydatagraphconvolutionalnetworksmulti-tasklearningtransformerspatiotemporalGoogleepidemicmodeling

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper sets out to determine whether two popular ingredients in epidemic-forecasting models—mobility data and graph convolutional layers—actually improve predictions of daily COVID-19 incidence. Across 28 EU countries and 49 US states, with a transformer as the temporal backbone, it finds no statistically significant gains from adding mobility data or from adding a GCN module with static, dynamic, or both adjacency matrices. The positive result is that incorporating mortality and hospitalization rates through multitask learning consistently improves forecast accuracy, suggesting that clinical surveillance data can stand in for explicit mobility information. A secondary observation ties the attention-derived dynamic spatial maps to lockdown periods, hinting that these maps may be usable as mobility indicators even when they do not help forecasts.

What carries the argument

The load-bearing mechanism is a transformer encoder-decoder backbone augmented with a graph convolutional module. Spatial structure enters through two adjacency matrices: a static physical-distance matrix built from a thresholded Gaussian kernel on site coordinates, and a dynamic matrix produced by a spatial attention head with a hard truncating constraint that sparsifies weak weights. A multitask-learning setup feeds incidence, mortality, hospitalization, and, in the pilot study, mobility rates as supplementary variables. The comparison logic is a one-sided Student t-test over five random seeds and progressive folds, with MAE and RMSE as the outcome metrics.

What would settle it

A replication using the same transformer backbone and data but with, say, fifty random seeds and per-architecture hyperparameter search that yields a significant MAE or RMSE decrease for the GCN-augmented models over the vanilla transformer would refute the claim that GCN modules lack evidence of contribution.

Watch

Extended reading notes

Core claim

The paper's central claim is that, on average, the GCN module with either a physical-distance adjacency matrix, an adaptively generated dynamic matrix, or both, lacks evidence of improving transformer-based COVID-19 incidence forecasting; reported one-sided t-tests yield P values near 0.5 for the MAE and RMSE comparisons. In the pilot study, adding mobility data to incidence data also fails to show consistent improvement, whereas replacing mobility with mortality and hospitalization data yields significantly lower errors in both regions. The paper interprets this as evidence that mobility information can be effectively mined from clinical outcomes, and that temporal information dominates spatial graph structure in this forecasting setting.

Load-bearing premise

The negative conclusions rest on the assumption that five random seeds, five progressive folds, and shared hyperparameters give the experiments enough statistical power to detect any practically meaningful improvement; if the design is underpowered, the insignificant P values do not rule out real gains.

Editorial extensions

If this is right

Adding mobility data to incidence data does not reliably improve short- or long-term COVID-19 case forecasts.
Mortality and hospitalization rates, used as auxiliary tasks, improve accuracy more than mobility data does and can act as a substitute source of mobility information.
GCN modules offer no significant benefit over a vanilla transformer for this task, and combining static and dynamic adjacency matrices tends to worsen MAE.
Attention-derived dynamic spatial maps correlate with lockdown periods in the EU and may serve as indicators of human movement restrictions.
Spatial modeling may still help in harder settings—longer output horizons and early, data-scarce folds—though the gains are not statistically significant.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the null result holds beyond this setup, forecasters should weight clinical surveillance data and parsimonious temporal models more heavily than mobility feeds or graph layers.
The paper's 'no evidence' wording leaves open the possibility of small real effects; a replication with more seeds or per-architecture hyperparameter tuning could detect them.
Because the dynamic maps track lockdown timing, they may be useful as policy surveillance tools even where they do not improve prediction.
A natural next test is whether other mobility sources, such as cell-phone location records, change the pilot-study verdict; the paper only assessed one mobility dataset.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: this is a competent head-to-head evaluation of mobility data and GCN modules on a transformer backbone for COVID-19 incidence forecasting across the EU and US. The strongest result is the pilot study: adding mortality and hospitalization data (IMH) consistently beats incidence-only and incidence+mobility, with significant gains in both regions. That finding looks real and is worth taking seriously. The second result—that GCN modules with static or dynamic adjacency matrices do not help—is plausible but not demonstrated as a true null. All the t-tests are one-sided over five random seeds, with no power analysis or confidence intervals. The paper itself says 'lacks evidence,' which is defensible, but the abstract says 'do not significantly enhance,' which overstates what an underpowered experiment can support. A true effect of a few percent MAE would frequently produce P-values in the 0.4–0.5 range with N=5. So the negative conclusion should be framed as 'we could not detect a benefit' rather than 'there is no benefit.' The lockdown-correlation analysis is the weakest part: it is qualitative, depends on post hoc threshold choices, and is partly circular because the dynamic maps come from models trained on data spanning the same lockdown periods. I would treat that as illustrative, not evidence. The missing citations for the transformer and GCN are minor but sloppy. On the plus side, the design is mostly clean: held-out test folds, progressive splits, external validation across two regions, and code and data are released. The sensitivity analysis with ARIMA, tree models, and DLinear is useful context. This paper deserves a serious referee: the empirical comparison is exactly the kind of negative result the field needs, and the IMH finding is actionable. The main revision request should be to add a power analysis or equivalence bounds, and to fix the abstract and discussion so they do not claim more than the statistics support. Yes, send it to review.

Referee Report

4 major / 5 minor

Summary. This paper evaluates whether mobility data and graph convolutional modules improve COVID-19 incidence forecasting across EU and US regions. In a pilot study, Google mobility data did not significantly improve a transformer-based model, whereas adding mortality and hospitalization data did. In main experiments, adding a GCN with physical-distance and/or attention-derived adjacency matrices did not significantly improve MAE/RMSE over a vanilla transformer under 5-fold progressive cross-validation and five random seeds. The authors additionally analyze attention-derived dynamic maps, introducing a mobility indicator Π, and report correlations with lockdown orders. Sensitivity analyses compare against ARIMA, tree-based models, and DLinear.

Significance. If the negative results are robust, they provide a useful cautionary finding for epidemic forecasting practice, suggesting that mobility feeds and GCN modules may add little over strong temporal baselines and that clinical surveillance data are more valuable. The study is notable for external validation on two distinct regions, held-out test folds, and public code and data. The main limitation is that the central negative claim is supported only by null-hypothesis tests without a power or equivalence analysis; additionally, the mobility-indicator analysis contains a formula/reporting inconsistency and a potential circularity due to training-set overlap.

major comments (4)

[Main experiments: Contribution of the GCN module; abstract] The conclusion that GCN modules "do not significantly enhance" forecasting is an absence-of-evidence claim. All comparisons are one-sided paired t-tests over N=5 random seeds, and the paper reports no confidence intervals, effect-size bounds, or power analysis. For example, the EU Trans+Adp MAE improvement of -1.83% at output length 12 carries P=0.441, and the pooled fold-1 improvement of -2.85% carries P=0.369; these P-values are exactly what an underpowered study would produce for a modest true effect. Please either add a non-inferiority/equivalence analysis with pre-specified equivalence margins, report effect-size confidence intervals, or explicitly state in the abstract and discussion that the evidence only warrants "no significant improvement detected," not "no improvement."
[Correlation between dynamic maps and human movement restrictions, Eq. (3)] The mobility indicator Π is defined with the indicator function I(M_n,(i,j)) applied to the attention score matrix M_n. Since M_n is a softmax-normalized matrix (Eqs. (9)-(10)), every element is strictly positive and Π would equal 1 for all maps. The reported values (e.g., Π=44.40% for EU and 9.48% for US) must have been computed on the thresholded/sparsified adjacency matrix A_s from Eq. (11). Please correct Eq. (3) and ensure all quantitative claims about Π refer to the intended matrix; as written, the formula is inconsistent with the reported numbers.
[Correlation between dynamic maps and human movement restrictions] The paper states that "we analyze all available maps, even if lockdown periods overlap with the training and validation sets." Because the dynamic maps are outputs of models trained on data spanning the same lockdown periods, the observed association between Π and lockdown orders could reflect memorization of training labels rather than a learned, generalizable sensitivity to mobility restrictions. Please retrain the map-generating models on pre-lockdown data only (or otherwise exclude lockdown periods from training) to establish that the maps track mobility restrictions out-of-sample, or temper the corresponding claims.
[Pilot study; Discussion] The claim that "mobility information can be effectively mined from mortality and hospitalization data" is not directly established by the experiments. The IMH input adds mortality and hospitalization time series, which are clinically informative for future incidence through direct temporal pathways; the paper does not include any control or mediation analysis to attribute the improvement to recovered mobility information. Please either soften this causal claim or add an analysis (e.g., an ablation using shuffled clinical data or an intermediate mobility prediction task) that isolates the mobility component.

minor comments (5)

[Perfomrance analysis for different lengths (heading)] The section heading "Perfomrance analysis for different lengths" contains a typo and should read "Performance."
[Methods, Algorithm 1] In Algorithm 1, line 3 sets ρ equal to the density of M_n when Ag exists, which would always be 1 for a softmax-normalized matrix; this contradicts the surrounding text ("if geographical adj. matrix exists, its sparsity is used"). Please correct the pseudocode.
[Methods, ARIMA and Linear Layer Deep Learning] The authors write that the AutoArima function comes from the "scikit-learn" package; AutoARIMA is not part of scikit-learn but is available in pmdarima or sktime (the latter is correctly cited in the Data Availability section).
[Correlation between dynamic maps and human movement restrictions] The text refers to "Supplementary Table 2 and Table 3" without distinguishing which numbers belong to which table; please make the citations specific and consistent with the supplement numbering.
[Figure 3 caption] The caption states the heatmaps are from "Trans+Adp" without specifying which output length or data split produced them; please clarify because the values of Π are not defined in the figure.

Circularity Check

1 steps flagged · score 2.0 of 10

No significant circularity: the central held-out forecasting comparisons are independent; only the secondary lockdown-map correlation uses in-sample training-period maps, as the paper discloses.

other [Results, section 'Correlation between dynamic maps and human movement restrictions']
"Since most lockdown orders occur early in the pandemic, we analyze all available maps, even if lockdown periods overlap with the training and validation sets."

The mobility indicator Π is computed from dynamic maps M_n produced by Trans+Adp, whose weights were trained on IMH data spanning the same calendar period as the lockdown orders. Comparing Π before, during, and after lockdown is therefore an in-sample association between model internals and training-period labels, not an out-of-sample prediction of mobility from the maps. The abstract's claim that spatial maps are 'sensitive indicators for mobility' is partly constructed by the model's exposure to lockdown-period data rather than established by an independent test. This is a disclosed secondary analysis and does not affect the main held-out-fold forecasting comparisons, which remain non-circular.

full rationale

The paper's primary derivation chain is self-contained: the GCN contribution is evaluated on temporally held-out test folds against a vanilla transformer, with results averaged over five random seeds, and no parameter is fitted to the test folds and then renamed as a prediction. The pilot study's IMH-versus-I comparison is likewise an out-of-sample forecasting comparison. No load-bearing self-citations or imported uniqueness theorems appear. The only partially circular element is the lockdown-map correlation, where the dynamic attention maps come from models trained on the same time span as the lockdowns; the paper explicitly acknowledges this overlap. Because that analysis is a secondary, exploratory claim rather than the central forecasting conclusion, the overall circularity score is low. The absence-of-evidence interpretation of the null P-values is a statistical-power concern, not a circularity concern.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The paper is an empirical evaluation, so there are no derived constants in the mathematical sense. The main free parameters are the truncation thresholds and sparsity ratios that control the dynamic adjacency matrices and the mobility indicator Pi. The load-bearing domain assumptions are that distance, Google mobility data, and lockdown durations are valid proxies for human movement, and that imputation does not bias the clinical data. Pi is an invented quantity with no independent validation.

free parameters (3)

Truncating threshold tau for sparsified adjacency matrices = 0.0357 (EU), 0.0179 (US)
Hand-set thresholds used to compute the mobility indicator Pi and to sparsify dynamic maps. They are chosen for forecasting performance, not for measuring mobility, so the lockdown-correlation analysis inherits this choice.
Sparsity ratio rho = 0.001 (initiation)
Hyper-parameter in Algorithm 1 controlling the density of generated adjacency matrices when no geographic matrix exists. It directly affects the dynamic maps used in the mobility indicator analysis.
Output length for lockdown analysis = 36 steps
Selected post hoc because the GCN module showed minor improvement at this horizon, which privileges the claimed correlation between dynamic maps and lockdowns.

assumptions (4)

domain assumption Euclidean distance between site coordinates is inversely related to travel likelihood.
Used to build the static physical-distance adjacency matrix Ag in Eq. (8). If distance does not proxy mobility, the static GCN map is misspecified.
domain assumption Google workplace mobility data is a valid proxy for human movement.
The pilot study and discussion rely on this to test mobility's contribution; the paper cites refs 18 and 21 but does not independently validate the proxy.
domain assumption Lockdown order duration is a proxy for human movement restrictions.
The correlation analysis uses lockdown durations as ground truth for mobility restrictions. The paper acknowledges this is a proxy, but the entire lockdown-correlation claim depends on it.
domain assumption Missing data imputation does not materially bias forecasting results.
Hospitalization data have many missing values and are imputed; the authors note imputation may shift data distributions (Discussion). This affects both the IMH results and the lockdown analysis.

invented entities (1)

Mobility indicator Pi
purpose: Ratio of non-zero elements in dynamic attention adjacency matrices, used to compare with lockdown orders and to claim that dynamic spatial maps are sensitive mobility indicators.
Pi is defined by the authors in Eq. (3) and is not calibrated against any independent mobility measurement. Its interpretation as a mobility indicator rests on attention maps produced by the forecasting model, whose thresholds were tuned for forecasting, not for measuring mobility.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Epidemic Forecasting: Evaluating the Role of Mobility Data and Graph Convolutional Networks." pith.science (2026). https://pith.science/paper/3ZVPQRWD

@misc{pith2026250611028,
  author       = {Pith},
  title        = {Pith review of: Enhancing Epidemic Forecasting: Evaluating the Role of Mobility Data and Graph Convolutional Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3ZVPQRWD}},
  note         = {Machine review of arXiv:2506.11028}
}

read the original abstract

Accurate prediction of contagious disease outbreaks is vital for informed decision-making. Our study addresses the gap between machine learning algorithms and their epidemiological applications, noting that methods optimal for benchmark datasets often underperform with real-world data due to difficulties in incorporating mobility information. We adopt a two-phase approach: first, assessing the significance of mobility data through a pilot study, then evaluating the impact of Graph Convolutional Networks (GCNs) on a transformer backbone. Our findings reveal that while mobility data and GCN modules do not significantly enhance forecasting performance, the inclusion of mortality and hospitalization data markedly improves model accuracy. Additionally, a comparative analysis between GCN-derived spatial maps and lockdown orders suggests a notable correlation, highlighting the potential of spatial maps as sensitive indicators for mobility. Our research offers a novel perspective on mobility representation in predictive modeling for contagious diseases, empowering decision-makers to better prepare for future outbreaks.

Figures

Figures reproduced from arXiv: 2506.11028 by the authors.

**Figure 1.** Research Pipeline The project consists of four major steps, termed “preprocess”, “train”, “evaluate” and “infer”. The first three steps serve as pipelines to answer the questions posed in the “infer” step. Results Research Pipeline Our study aims to find the core contributor to the spatiotemporal neural networks for integrating mobility information into SARS-CoV-2 incidence forecasting. This objective is divided int… view at source ↗

**Figure 2.** MAE and RMSE scores for incidence forecasting during the pilot study, spanning from July 2, 2022, to November 15, 2022. The correspondence between input and output lengths is maintained at 12 steps each. we propose an alternative hypothesis suggesting that mortality and hospitalization data alone are not adequate sources of mobility information, which implies that the incorporation of mobility data is not redundant … view at source ↗

**Figure 3.** Heatmaps of “Trans+Adp” dynamic maps from the pilot datasets of the EU and US. Moving from left to right, the datasets used are I, IB, IMH, and IBMH. summarized in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (5 more)

**Figure 4.** Figure 4: Network Structure of “Trans+GCN+Adp” Performance analysis under varying data availability In [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Incidence forecasting MAE scores for test sets of 5-fold split and the final test set. The rows represent the output lengths, 3,6,12,24,36. The columns represent folds. Supplementary [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

**Figure 6.** Figure 6: Incidence forecasting RMSE scores for test sets of 5-fold split and the final test set. The rows represent the output lengths, 3,6,12,24,36. The columns represent folds. of site selection and the potential influence of regional characteristics on model performance cann…

**Figure 7.** Figure 7: Heatmaps of most connected dynamic maps. Moving from left to right, EU Trans+Adp, EU Trans+GCN+Adp, US Trans+Adp, US Trans+GCN+Adp. reducing mobility38. Research indicates that lifting stringent NPIs, like lockdown measures, requires coordinated efforts among closely c…

**Figure 8.** Figure 8: MAE and RMSE scores of incidence forecasting for EU and US, including ARIMA, Tree-based models, DLinear, and Transformer. Results on output length of 12 are reported. 11/19 [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 46 canonical work pages

[1]

Elliott, P. et al. Exponential growth, high prevalence of SARS-CoV-2, and vaccine effectiveness associated with the Delta variant. Science 374, eabl9551 (2021)

work page 2021
[2]

& Vespignani, A

Colizza, V ., Barrat, A., Barthélemy, M. & Vespignani, A. The role of the airline transportation network in the prediction and predictability of global epidemics. PNAS 103, 2015–2020 (2006)

work page 2006
[3]

Halloran, M. E. et al. Ebola: Mobility data. Science 346, 433–433 (2014)

work page 2014
[4]

& Helbing, D

Brockmann, D. & Helbing, D. The Hidden Geometry of Complex, Network-Driven Contagion Phenomena. Science 342, 1337–1342 (2013)

work page 2013
[5]

Balcan, D. et al. Multiscale mobility networks and the spatial spreading of infectious diseases. PNAS 106, 21484–21489 (2009)

work page 2009
[6]

Jia, J. S. et al. Population flow drives spatio-temporal distribution of COVID-19 in China. Nature 582, 389–394 (2020)

2020
[7]

& Wang, X

Abdelrahman, Z., Li, M. & Wang, X. Comparative Review of SARS-CoV-2, SARS-CoV, MERS-CoV, and Influenza A Respiratory Viruses. Front. Immunol. 11, 552909 (2020)

work page 2020
[8]

& Bogoch, I

Findlater, A. & Bogoch, I. I. Human Mobility and the Global Spread of Infectious Diseases: A Focus on Air Travel. Trends Parasitol. 34, 772–783 (2018)

work page 2018

Show all 47 references

[9]

Wilson, M. E. & Chen, L. H. Travellers give wings to novel coronavirus (2019-nCoV). J Travel. Med. 27, taaa015 (2020)

2020
[10]

& Shi, X

Chen, S., Li, Q., Gao, S., Kang, Y . & Shi, X. State-specific projection of COVID-19 infection in the United States and evaluation of three major control measures. Sci. Rep. 10, 22429 (2020)

2020
[11]

Ruktanonchai, N. W. et al. Assessing the impact of coordinated COVID-19 exit strategies across Europe. Science 369, 1465–1470 (2020)

2020
[12]

Meloni, S. et al. Modeling human mobility responses to the large-scale spreading of infectious diseases. Sci. Rep. 1, 62 (2011)

2011
[13]

Sills, J. et al. Aggregated mobility data could help fight COVID-19. Science 368, 145–146 (2020)

2020
[14]

Kraemer, M. U. G. et al. The effect of human mobility and control measures on the COVID-19 epidemic in China. Science 368, 493–497 (2020)

2020
[15]

& Klein, E

Gatalo, O., Tseng, K., Hamilton, A., Lin, G. & Klein, E. Associations between phone mobility data and COVID-19 cases. Lancet Infect Dis. 21, e111 (2021)

2021
[16]

& De Angelis, D

Birrell, P., Blake, J., van Leeuwen, E., Gent, N. & De Angelis, D. Real-time nowcasting and forecasting of COVID-19 dynamics in England: The first wave. Phil. Trans. R. Soc. B 376, 20200279
[17]

Scientific evidence supporting the gov- ernment response to coronavirus (COVID-19)

Scientific Advisory Group for Emergencies (SAGE). Scientific evidence supporting the gov- ernment response to coronavirus (COVID-19). https://www.gov.uk/government/collections/ scientific-evidence-supporting-the-government-response-to-coronavirus-covid-19
[18]

& Chollet, F

Ward, T., Johnsen, A., Ng, S. & Chollet, F. Forecasting SARS-CoV-2 transmission and clinical risk at small spatial scales by the application of machine learning architectures to syndromic surveillance data. Nat. Mach. Intell. 4, 814–827 (2022)

2022
[19]

Reduction in mobility and COVID-19 transmission

Nouvellet, P.et al. Reduction in mobility and COVID-19 transmission. Nat. Commun. 12, 1090 (2021)

2021
[20]

Badr, H. S. et al. Association between mobility patterns and COVID-19 transmission in the USA: A mathematical modelling study. Lancet Infect Dis. 20, 1247–1254 (2020)

2020
[21]

Liu, Y .et al. Associations between changes in population mobility in response to the COVID-19 pandemic and socioeco- nomic factors at the city level in China and country level worldwide: A retrospective, observational study. The Lancet Digit. Heal. 3, e349–e359 (2021)

2021
[22]

Nixon, K. et al. Real-time COVID-19 forecasting: Challenges and opportunities of model performance and translation. Lancet Digit. Heal. 4, e699–e701 (2022)

2022
[23]

Jia, W. et al. Extended SIR Prediction of the Epidemics Trend of COVID-19 in Italy and Compared With Hunan, China. Front. Med. 7, 169 (2020)

2020
[24]

& Golabpour, A

Ahouz, F. & Golabpour, A. Predicting the incidence of COVID-19 using data mining. BMC Public Heal. 21, 1087 (2021)

2021
[25]

Zhao, D., Zhang, R., Zhang, H. & He, S. Prediction of global omicron pandemic using ARIMA, MLR, and Prophet models. Sci. Rep. 12, 18138 (2022). 17/19

2022
[26]

M., Rosete-Verges, F

Franch-Pardo, I., Napoletano, B. M., Rosete-Verges, F. & Billa, L. Spatial analysis and GIS in the study of COVID-19. A review. Sci. Total. Environ.739, 140033 (2020)

2020
[27]

& Ning, Y

Deng, S., Wang, S., Rangwala, H., Wang, L. & Ning, Y . Cola-GNN: Cross-location attention based graph neural networks for long-term ILI prediction. In Proc. 29th ACM CIKM International Conference on Information and Knowledge Management, 245–254 (2020)

2020
[28]

Gao, J. et al. STAN: Spatio-temporal attention network for pandemic prediction using real-world evidence. JAMIA 28, 733–743 (2021)

2021
[29]

& Yan, X

Jin, X., Wang, Y .-X. & Yan, X. Inter-series attention model for COVID-19 forecasting. InProc. 2021 SIAM International Conference on Data Mining, 495–503 (SIAM, 2021)

2021
[30]

Gao, J. et al. Evidence-driven spatiotemporal COVID-19 hospitalization prediction with Ising dynamics. Nat. Commun. 14, 3093 (2023)

2023
[31]

Multitask learning

Caruana, R. Multitask learning. Mach. Learn. 28, 41–75 (1997)

1997
[32]

& Yang, Q

Zhang, Y . & Yang, Q. A survey on multi-task learning.IEEE Trans. Knowl. Data Eng. 34, 5586–5609 (2022)

2022
[33]

Kavouras, I. et al. COVID-19 Spatio-Temporal Evolution Using Deep Learning at a European Level. Sensors 22, 3658 (2022)

2022
[34]

Pandemic and lockdown: A territorial approach to COVID-19 in China, Italy and the United States

Ren, X. Pandemic and lockdown: A territorial approach to COVID-19 in China, Italy and the United States. Eurasian Geogr. Econ. 61, 423–434 (2020)

2020
[35]

& Neumayer, E

Plümper, T. & Neumayer, E. Lockdown policies and the dynamics of the first wave of the Sars-CoV-2 pandemic in Europe. J. Eur. Public Policy29, 321–341 (2022)

2022
[36]

Brauner, J. M. et al. Inferring the effectiveness of government interventions against COVID-19. Science 371, eabd9338 (2021)

2021
[37]

Lison, A. et al. Effectiveness assessment of non-pharmaceutical interventions: Lessons learned from the COVID-19 pandemic. Lancet Public Heal. 8, e311–e317 (2023)

2023
[38]

T., Burger, M., Sun, S., Dobson, R

Snoeijer, B. T., Burger, M., Sun, S., Dobson, R. J. B. & Folarin, A. A. Measuring the effect of Non-Pharmaceutical Interventions (NPIs) on mobility during the COVID-19 pandemic using global mobility data. npj Digit. Med. 4, 1–12 (2021)

2021
[39]

Ruktanonchai, N. W. et al. Assessing the impact of coordinated covid-19 exit strategies across europe. Science 369, 1465–1470 (2020)

2020
[40]

Threat assessment brief: Implications of the further emergence and spread of the SARS-CoV-2 B.1.1.529 variant of concern (Omicron) for the eu/eea - first update

European Centre for Disease Prevention and Control. Threat assessment brief: Implications of the further emergence and spread of the SARS-CoV-2 B.1.1.529 variant of concern (Omicron) for the eu/eea - first update. Tech. Rep. (2021). https://www.ecdc.europa.eu/en/publications-d...

2021
[41]

Omicron surge is ‘unlike anything we’ve ever seen,’ expert says

CNN Health. Omicron surge is ‘unlike anything we’ve ever seen,’ expert says. https://www.cnn.com/2021/12/30/health/ us-coronavirus-thursday/index.html

2021
[42]

& Liu, Y

Li, Y ., Yu, R., Shahabi, C. & Liu, Y . Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting. In Proc 6th International Conference on Learning Representations (OpenReview.net, 2017)

2017
[43]

& Guestrin, C

Chen, T. & Guestrin, C. XGBoost: A scalable tree boosting system. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (ACM, 2016)

2016
[44]

Ke, G. et al. LightGBM: A highly efficient gradient boosting decision tree. In Advances in Neural Information Processing Systems 30 (Curran Associates, 2017)

2017
[45]

Random Forests

Breiman, L. Random Forests. Mach. Learn. 45, 5–32 (2001)

2001
[46]

Zeng, A., Chen, M., Zhang, L. & Xu, Q. Are transformers effective for time series forecasting? In Proc 37th AAAI Conference on Artificial Intelligence (AAAI Press, 2023)

2023
[47]

& Gardner, L

Dong, E., Du, H. & Gardner, L. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect. Dis. 20, 533–534 (2020). Acknowledgements This work is supported in part by the National Natural Science Foundation of China under Grant Nos. 62276127. 18/19 Author...

2020

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Elliott, P. et al. Exponential growth, high prevalence of SARS-CoV-2, and vaccine effectiveness associated with the Delta variant. Science 374, eabl9551 (2021)

work page 2021

[2] [2]

& Vespignani, A

Colizza, V ., Barrat, A., Barthélemy, M. & Vespignani, A. The role of the airline transportation network in the prediction and predictability of global epidemics. PNAS 103, 2015–2020 (2006)

work page 2006

[3] [3]

Halloran, M. E. et al. Ebola: Mobility data. Science 346, 433–433 (2014)

work page 2014

[4] [4]

& Helbing, D

Brockmann, D. & Helbing, D. The Hidden Geometry of Complex, Network-Driven Contagion Phenomena. Science 342, 1337–1342 (2013)

work page 2013

[5] [5]

Balcan, D. et al. Multiscale mobility networks and the spatial spreading of infectious diseases. PNAS 106, 21484–21489 (2009)

work page 2009

[6] [6]

Jia, J. S. et al. Population flow drives spatio-temporal distribution of COVID-19 in China. Nature 582, 389–394 (2020)

2020

[7] [7]

& Wang, X

Abdelrahman, Z., Li, M. & Wang, X. Comparative Review of SARS-CoV-2, SARS-CoV, MERS-CoV, and Influenza A Respiratory Viruses. Front. Immunol. 11, 552909 (2020)

work page 2020

[8] [8]

& Bogoch, I

Findlater, A. & Bogoch, I. I. Human Mobility and the Global Spread of Infectious Diseases: A Focus on Air Travel. Trends Parasitol. 34, 772–783 (2018)

work page 2018