Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
157 tokens/sec
GPT-4o
43 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Forecasting trends in food security with real time data (2312.00626v3)

Published 1 Dec 2023 in cs.LG, physics.soc-ph, and stat.ML

Abstract: Early warning systems are an essential tool for effective humanitarian action. Advance warnings on impending disasters facilitate timely and targeted response which help save lives and livelihoods. In this work we present a quantitative methodology to forecast levels of food consumption for 60 consecutive days, at the sub-national level, in four countries: Mali, Nigeria, Syria, and Yemen. The methodology is built on publicly available data from the World Food Programme's global hunger monitoring system which collects, processes, and displays daily updates on key food security metrics, conflict, weather events, and other drivers of food insecurity. In this study we assessed the performance of various models including Autoregressive Integrated Moving Average (ARIMA), Extreme Gradient Boosting (XGBoost), Long Short Term Memory (LSTM) Network, Convolutional Neural Network (CNN), and Reservoir Computing (RC), by comparing their Root Mean Squared Error (RMSE) metrics. Our findings highlight Reservoir Computing as a particularly well-suited model in the field of food security given both its notable resistance to over-fitting on limited data samples and its efficient training capabilities. The methodology we introduce establishes the groundwork for a global, data-driven early warning system designed to anticipate and detect food insecurity.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (68)
  1. Jagtap, S. et al. The russia-ukraine conflict: Its implications for the global food supply chains. \JournalTitleFoods 11, DOI: 10.3390/foods11142098 (2022).
  2. Impacts of the russia-ukraine war on global food security: Towards more sustainable and resilient food systems? \JournalTitleFoods 11, DOI: 10.3390/foods11152301 (2022).
  3. Alexander, P. et al. High energy and fertilizer prices are more damaging than food export curtailment from ukraine and russia for food prices, health and the environment. \JournalTitleNature Food 1–12 (2022).
  4. The impact of covid-19 on diet quality, food security and nutrition in low and middle income countries: A systematic review of the evidence. \JournalTitleClinical Nutrition 41, 2955–2964 (2022).
  5. WFP. War in ukraine drives global food crisis (2022). https://www.wfp.org/publications/war-ukraine-drives-global-food-crisis.
  6. WFP. Global report on food crises - 2022 (2022). https://www.wfp.org/publications/global-report-food-crises-2022.
  7. Time-series forecasting with deep learning: a survey. \JournalTitlePhilosophical Transactions of the Royal Society A 379, 20200209 (2021).
  8. Computer aided system of time series analysis methods for forecasting the epidemics outbreaks. In 2019 IEEE 15th International Conference on the Experience of Designing and Application of CAD Systems (CADSM), 1–4 (IEEE, 2019).
  9. Kraemer, M. U. et al. Past and future spread of the arbovirus vectors aedes aegypti and aedes albopictus. \JournalTitleNature microbiology 4, 854–863 (2019).
  10. Tizzoni, M. et al. Real-time numerical forecast of global epidemic spreading: case study of 2009 a/h1n1pdm. \JournalTitleBMC medicine 10, 1–31 (2012).
  11. On the predictability of infectious disease outbreaks. \JournalTitleNature communications 10, 1–8 (2019).
  12. Ardabili, S. F. et al. Covid-19 outbreak prediction with machine learning. \JournalTitleAlgorithms 13, 249 (2020).
  13. Analysis, prediction and evaluation of covid-19 datasets using machine learning algorithms. \JournalTitleInternational Journal 8, 2199–2204 (2020).
  14. Timmermann, A. Forecasting methods in finance. \JournalTitleAnnual Review of Financial Economics 10, 449–479 (2018).
  15. Forecasting in economics and finance. \JournalTitleAnnual Review of Economics 8, 81–110 (2016).
  16. Financial time series forecasting with deep learning: A systematic literature review: 2005–2019. \JournalTitleApplied soft computing 90, 106181 (2020).
  17. Universal features of price formation in financial markets: perspectives from deep learning. \JournalTitleQuantitative Finance 19, 1449–1459 (2019).
  18. A review on time series forecasting techniques for building energy consumption. \JournalTitleRenewable and Sustainable Energy Reviews 74, 902–924 (2017).
  19. Schultz, M. G. et al. Can deep learning beat numerical weather prediction? \JournalTitlePhilosophical Transactions of the Royal Society A 379, 20200097 (2021).
  20. Mudelsee, M. Trend analysis of climate time series: A review of methods. \JournalTitleEarth-science reviews 190, 310–322 (2019).
  21. On the forecastability of food insecurity. \JournalTitleScientific Reports 13, 2793 (2023).
  22. Martini, G. et al. Machine learning can guide food security efforts when primary data are not available. \JournalTitleNature Food 3, 716–728 (2022).
  23. Predicting food crises using news streams. \JournalTitleScience Advances 9, eabm3449 (2023).
  24. Organization, W. H. et al. The state of food security and nutrition in the world 2020: transforming food systems for affordable healthy diets, vol. 2020 (Food & Agriculture Org., 2020).
  25. Refinements to the fao methodology for estimating the prevalence of undernourishment indicator. \JournalTitleDocumento de trabajo 2014, 14–05 (2014).
  26. Causal structure learning for famine prediction. In 2010 AAAI Spring Symposium Series (2010).
  27. Machine learning classification technique for famine prediction. In Proceedings of the world congress on engineering, vol. 2, 4–9 (2011).
  28. A data-driven approach improves food insecurity crisis prediction. \JournalTitleWorld Development 122, 399–409 (2019).
  29. Predicting food crises. \JournalTitleThe World Bank (2020).
  30. Stochastic modeling of food insecurity. \JournalTitleThe World Bank (2020).
  31. Westerveld, J. J. et al. Forecasting transitions in the state of food security with machine learning using transferable features. \JournalTitleScience of the Total Environment 786, 147366 (2021).
  32. Deléglise, H. et al. Food security prediction from heterogeneous data combining machine and deep learning methods. \JournalTitleExpert Systems with Applications 190, 116189 (2022).
  33. Anticipating drought-related food security changes. \JournalTitleNature Sustainability 5, 956–964 (2022).
  34. Transitions into and out of food insecurity: A probabilistic approach with panel data evidence from 15 countries. \JournalTitleWorld Development 159, 106035 (2022).
  35. WFP. Food security analysis (2022). https://www.wfp.org/food-security-analysis.
  36. WFP. Hungermap live (2022). https://hungermap.wfp.org.
  37. WFP. The world food programme’s real-time monitoring systems: Approaches and methodologies (2021). https://docs.wfp.org/api/documents/WFP-0000135070/download/.
  38. McBride, L. et al. Predicting poverty and malnutrition for targeting, mapping, monitoring, and early warning. \JournalTitleApplied Economic Perspectives and Policy 44, 879–892 (2022).
  39. Machine learning in agricultural economics. In Handbook of Agricultural Economics, vol. 5, 4551–4612 (Elsevier, 2021).
  40. Machine learning for food security: Principles for transparency and usability. \JournalTitleApplied Economic Perspectives and Policy 44, 893–910 (2022).
  41. Real-time computing without stable states: A new framework for neural computation based on perturbations. \JournalTitleNeural Computation 14, 2531–2560, DOI: 10.1162/089976602760407955 (2002).
  42. Jaeger, H. The “echo state” approach to analysing and training recurrent neural networks-with an erratum note. \JournalTitleBonn, Germany: German National Research Center for Information Technology GMD Technical Report 148, 13 (2001).
  43. Reservoir computing approaches to recurrent neural network training. \JournalTitleComputer Science Review 3, 127–149 (2009).
  44. Data-driven predictions of a multiscale lorenz 96 chaotic system using machine-learning methods: reservoir computing, artificial neural network, and long short-term memory network. \JournalTitleNonlinear Processes in Geophysics 27, 373–389, DOI: 10.5194/npg-27-373-2020 (2020).
  45. Vlachas, P. R. et al. Backpropagation algorithms and reservoir computing in recurrent neural networks for the forecasting of complex spatiotemporal dynamics (2019). 1910.05266.
  46. Prediction of chaotic time series using recurrent neural networks and reservoir computing techniques: A comparative study. \JournalTitleMachine learning with applications 8, 100300 (2022).
  47. Anticipating food price crises by reservoir computing. \JournalTitleAvailable at SSRN 4457897 (2023).
  48. Prediction and comparison of in-vehicle co2 concentration based on arima and lstm models. \JournalTitleApplied Sciences 13, 10858 (2023).
  49. Menculini, L. et al. Comparing prophet and deep learning to arima in forecasting wholesale food prices. \JournalTitleForecasting 3, 644–662 (2021).
  50. Arima vs lstm on nasdaq stock exchange data. \JournalTitleProcedia Computer Science 207, 3836–3845 (2022).
  51. Convolutional neural network–component transformation (cnn–ct) for confirmed covid-19 cases. \JournalTitleMathematical and Computational Applications 26, 29 (2021).
  52. Predicting high-dimensional heterogeneous time series employing generalized local states. \JournalTitlePhys. Rev. Research 3, 023215, DOI: 10.1103/PhysRevResearch.3.023215 (2021).
  53. Next generation reservoir computing. \JournalTitleNature communications 12, 1–8 (2021).
  54. Forecasting chaotic systems with very low connectivity reservoir computers. \JournalTitleChaos: An Interdisciplinary Journal of Nonlinear Science 29, 123108 (2019).
  55. Attractor reconstruction by machine learning. \JournalTitleChaos: An Interdisciplinary Journal of Nonlinear Science 28, 061104 (2018).
  56. Lu, Z. et al. Reservoir observers: Model-free inference of unmeasured variables in chaotic systems. \JournalTitleChaos: An Interdisciplinary Journal of Nonlinear Science 27, 041102 (2017).
  57. Breaking symmetries of the reservoir equations in echo state networks. \JournalTitleChaos: An Interdisciplinary Journal of Nonlinear Science 30, 123142 (2020).
  58. Ridge regression: Biased estimation for nonorthogonal problems. \JournalTitleTechnometrics 12, 55–67 (1970).
  59. Application of the arima model on the covid-2019 epidemic dataset. \JournalTitleData in brief 29, 105340 (2020).
  60. Long short-term memory. \JournalTitleNeural computation 9, 1735–1780 (1997).
  61. WFP. Seasonal explorer (2022). https://dataviz.vam.wfp.org/seasonal_explorer/rainfall_vegetation/visualizations.
  62. (CHC), C. H. C. Chirps: Rainfall estimates from rain gauge and satellite observations (2022). https://www.chc.ucsb.edu/data/chirps.
  63. NASA. Moderate resolution imaging spectroradiometer (modis) (2022). https://modis.gsfc.nasa.gov/.
  64. ACLED. The armed conflict location & event data project (2022). https://acleddata.com.
  65. Introducing acled: an armed conflict location and event dataset: special data feature. \JournalTitleJournal of peace research 47, 651–660 (2010).
  66. Alert at maradi: preventing food crises by using price signals. \JournalTitleWorld Development 40, 1882–1894 (2012).
  67. WFP. Calculation and use of the alert for price spikes (alps) indicator, technical guidance note, analysis and nutrition service (2014). https://documents.wfp.org/stellent/groups/public/documents/manual_guide_proced/wfp264186.pdf.
  68. WFP. Economic explorer (2022). https://dataviz.vam.wfp.org/economic_explorer/prices.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets