Tracking food insecurity from tweets using data mining techniques

Lukyamuzi, Andrew; Ngubiri, John; Okori, Washington

doi:10.1145/3195528.3195531

Cited by 6 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, this study improved upon previous work on food security topic modeling that used too broad or too narrow searches to retrieve all relevant food security tweets. 41,42 The current study https://doi.org/10.1017/dmp.2022.207 Published online by Cambridge University Press succeeded in defining first a search string, based on a literature review and stakeholder input.…”

Section: Discussionmentioning

confidence: 99%

“…Other fields have improved their real-time analyses by feeding their base topic model results into more advanced supervised learning algorithms and feature development. 41 Current platforms for showing food security information are lacking in real-time data, modern user interfaces, and actionable or usable data for policymakers and aid organization to make real-time resource allocation decisions.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing Artificial Intelligence for Twitter-based Public Discourse on Food Security During the COVID-19 Pandemic

Martin¹,

Poirier²,

Rosenblum

et al. 2022

Disaster med. public health prep.

View full text Add to dashboard Cite

Objective: Food security during public health emergencies relies on situational awareness of needs and resources. Artificial intelligence (AI) has revolutionized situational awareness during crises, allowing the allocation of resources to needs through machine learning algorithms. Limited research exists monitoring Twitter for changes in the food security-related public discourse during the COVID-19 pandemic. We aim to address that gap with AI by classifying food security topics on Twitter and showing topic frequency per day. Methods: Tweets were scraped from Twitter from January 2020 through December 2021 using food security keywords. Latent Dirichlet Allocation (LDA) topic modeling was performed, followed by time-series analyses on topic frequency per day. Results: 237,107 tweets were scraped and classified into topics, including food needs and resources, emergency preparedness and response, and mental/physical health. After the WHO’s pandemic declaration, there were relative increases in topic density per day regarding food pantries, food banks, economic and food security crises, essential services, and emergency preparedness advice. Threats to food security in Tigray emerged in 2021. Conclusions: AI is a powerful yet underused tool to monitor food insecurity on social media. Machine learning tools to improve emergency response should be prioritized, along with measurement of impact. Further food insecurity word patterns testing, as generated by this research, with supervised machine learning models can accelerate the uptake of these tools by policymakers and aid organizations.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Enhancing Artificial Intelligence for Twitter-based Public Discourse on Food Security During the COVID-19 Pandemic

Martin¹,

Poirier²,

Rosenblum

et al. 2022

Disaster med. public health prep.

View full text Add to dashboard Cite

show abstract

“…No study indicated that new algorithms were needed, created or tested. Some studies chose their algorithm from among the most popular or based on past evidence 77 and experience, while other studies chose their algorithm by testing of multiple algorithms within their research 73,78 to determine the best fit and/or accurate model.…”

Section: Major Technical Findings and Challengesmentioning

confidence: 99%

“…Two studies reported the lack of national data as a major setback in predicting food insecurity at a sub-national level. 73,78 It is unreasonable for single studies to collect the national-level data that is needed for some models to run accurately. They overcame this challenge through historical records from years where food security data was collected at a national level and use of available apps with large in-country user bases (e.g., OLIO and conducting cross-sectional household interviews to approximate the larger population).…”

Section: Major Technical Findings and Challengesmentioning

confidence: 99%

Harnessing Artificial Intelligence to Improve Food Assistance: A Scoping Review of Machine Learning Tools

Martin¹,

Sedoc²,

Poirier³

et al. 2022

Preprint

View full text Add to dashboard Cite

Background: Machine learning has revolutionized situational awareness during disaster management by classifying, clustering, and predicting impacted locations and people. Despite its importance, no review has been conducted on machine learning tools for food assistance efforts during emergency or non-emergency situations. The purpose of this scoping review is to address that gap. Methods: Keywords were defined within the concepts of food assistance and machine learning. After the database searches, PRISMA guidelines were followed to perform a partnered, two-round scoping literature review. Text mining and Latent Dirichlet Allocation topic modeling algorithms were used to determine trends. Results: 28 articles met criteria and were included in the analysis. The types of study designs included: model development (42.9%), non-study (i.e., text and opinion) (28.6%), qualitative research (14.3%), case study (10.7%), and meta-analysis (3.6%). There were no quantitative studies. The machine learning tools’ main functions were improving SNAP programs (32.1%), detecting needs and resources (25%), predicting food insecurity (21.4%), and situational awareness of current food insecurity issues (21.4%). None of these studies took place during a disaster or explicitly addressed emergency mitigation, preparedness, or recovery. All of the studies were in early phases of development and implementation. Conclusion: Machine learning tools for improving situational awareness, resource allocation, policymaking, and prediction have the potential to improve food assistance, but there is a lack of implementation and evaluation during all disaster phases. Also needed is more formative work on generating food-related queries and defining variables and features of food security.

show abstract

“…Machine Learning methods are increasingly used to extract relevant information from complex and heterogeneous FS-related data, and several studies have attempted to detect food insecurity and crisis using machine learning techniques [14,1,10] with encouraging but improving results. A group of machine learning methods called deep learning is increasingly being used and is very effective in analysing complex and heterogeneous data [7].…”

Section: Introductionmentioning

confidence: 99%

Linking Heterogeneous Data for Food Security Prediction

Deléglise

Bégué

Interdonato

et al. 2020

ECML PKDD 2020 Workshops

View full text Add to dashboard Cite

Identifying food insecurity situations timely and accurately is a complex challenge. To prevent food crisis and design appropriate interventions, several food security warning and monitoring systems are very active in food-insecure countries. However, the limited types of data selected and the limitations of data processing methods used make it difficult to apprehend food security in all its complexity. In this work, we propose models that aim to predict two key indicators of food security: the food consumption score and the household dietary diversity score. These indicators are time consuming and costly to obtain. We propose using heterogeneous data as explanatory variables that are more convenient to collect. These indicators are calculated using data from the permanent agricultural survey conducted by the Burkinabe government and available since 2009. The proposed models use deep and machine learning methods to obtain an approximation of food security indicators from heterogeneous explanatory data. The explanatory data are rasters (population densities, rainfall estimates, land use, etc.), GPS points (of hospitals, schools, violent events), quantitative economic variables (maize prices, World Bank variables), meteorological and demographic variables. A basic research issue is to perform pre-processing adapted to each type of data and then to find the right methods and spatio-temporal scale to combine them. This work may also be useful in an operational approach, as the methods discussed could be used by food security warning and monitoring systems to complement their methods to obtain estimates of key indicators a few weeks in advance and to react more quickly in case of famine.

show abstract

Tracking food insecurity from tweets using data mining techniques

Cited by 6 publications

References 13 publications

Enhancing Artificial Intelligence for Twitter-based Public Discourse on Food Security During the COVID-19 Pandemic

Enhancing Artificial Intelligence for Twitter-based Public Discourse on Food Security During the COVID-19 Pandemic

Harnessing Artificial Intelligence to Improve Food Assistance: A Scoping Review of Machine Learning Tools

Linking Heterogeneous Data for Food Security Prediction

Contact Info

Product

Resources

About