“…Seminal work by Lentz and collaborators addressed this challenge, obtaining predictions that explain up to 65% of the variation in food consumption, although limited to Malawi only 22 . Similar studies in the context of Ethiopia 23 and Burkina Faso 24 were also recently proposed. Here we make use of a unique dataset of sub-national-level food consumption and food-based coping data collected during the past 15 years across, respectively, 78 and 41 countries and aggregated by first-level administrative country sub-divisions (for example, departments, provinces and so on), allowing the development and validation of nowcasting predictive models of food security indicators on a global scale.…”
Section: Machine Learning Can Guide Food Security Efforts When Primar...supporting
haracterizing the socio-economic status of populations and providing reliable and up-to-date estimates of who the most vulnerable are, how many they are, where they live and why they are vulnerable is essential for governments and humanitarian organizations to make informed and timely decisions on implementation of humanitarian assistance policies and programmes 1 . These data are traditionally collected through face-to-face surveys. However, these are expensive, time-consuming and, in certain areas, not possible to perform due to conflict, disease, insecurity or remoteness. Therefore, during the past few years, researchers have begun to investigate the potential of non-traditional data and new computational methods to estimate vulnerabilities and socio-economic characteristics when primary data are not available. In these studies, mobile phone data 2 , satellite imagery 3 , a combination of both 4,5 , mobile money transaction records 6 , geolocated Wikipedia articles 7 or tweets 8 and social media advertising data 9 have been used in combination with state-of-the-art machine learning methods to provide reliable estimates of poverty at different spatial resolutions for several sub-Saharan African countries and southern and southeastern Asian ones.
“…Seminal work by Lentz and collaborators addressed this challenge, obtaining predictions that explain up to 65% of the variation in food consumption, although limited to Malawi only 22 . Similar studies in the context of Ethiopia 23 and Burkina Faso 24 were also recently proposed. Here we make use of a unique dataset of sub-national-level food consumption and food-based coping data collected during the past 15 years across, respectively, 78 and 41 countries and aggregated by first-level administrative country sub-divisions (for example, departments, provinces and so on), allowing the development and validation of nowcasting predictive models of food security indicators on a global scale.…”
Section: Machine Learning Can Guide Food Security Efforts When Primar...supporting
haracterizing the socio-economic status of populations and providing reliable and up-to-date estimates of who the most vulnerable are, how many they are, where they live and why they are vulnerable is essential for governments and humanitarian organizations to make informed and timely decisions on implementation of humanitarian assistance policies and programmes 1 . These data are traditionally collected through face-to-face surveys. However, these are expensive, time-consuming and, in certain areas, not possible to perform due to conflict, disease, insecurity or remoteness. Therefore, during the past few years, researchers have begun to investigate the potential of non-traditional data and new computational methods to estimate vulnerabilities and socio-economic characteristics when primary data are not available. In these studies, mobile phone data 2 , satellite imagery 3 , a combination of both 4,5 , mobile money transaction records 6 , geolocated Wikipedia articles 7 or tweets 8 and social media advertising data 9 have been used in combination with state-of-the-art machine learning methods to provide reliable estimates of poverty at different spatial resolutions for several sub-Saharan African countries and southern and southeastern Asian ones.
“…HDDS measures food consumption frequency and diversity by focusing on the nutritional quality of the diet, and it is calculated based on the number of different food groups consumed in the last 24 hours. We calculated the values of FCS and H DDS by using data from the permanent agricultural survey conducted by the Burkinabè government, that was kindly provided to us by the Ministry of Agriculture of Burkina Faso, as for the work in [4].…”
Section: T Xt -Fs Compared To Survey Based Indicatorsmentioning
confidence: 99%
“…Some examples may be spatial information (e.g., population density, land use, soil quality), volunteered geographical information (number of hospitals and schools, number and details about violent events) and economic indicators (e.g., price of representing goods). Recent literature has shown how these heterogeneous data can be exploited to predict food security indicators through advanced data science methods [4,11,21], e.g., multi-branch neural networks able to integrate data of different types and at different scales, by also taking into account spatial and temporal context. Nevertheless, while the performance of these approaches seem to be promising, they are still far from being optimal, and strongly dependent from the study area taken into account (e.g., availability and quality of the data may not be the same over different country, as well as the correlation with food security indicators).…”
Food security is a major concern in many countries all over the world. After a relatively long period characterized by a positive trend, the number and severity of food insecurity situations has been growing again in recent years, with alarming projections for the near future. While several Early Warning Systems (EWS) exist to monitor this phenomenon and guide the interventions of governments and ONGs, such systems rely on a narrow set of data types, i.e., mainly satellite imagery and survey data. These data can explain just a limited number of the multiple factors that impact on food security, thus producing an incomplete picture of the real scenario. In this work, we propose a spatio-temporal analysis of unconventional textual data (i.e., YouTube transcriptions and articles from local news papers) to support the explanatory process of food insecurity situations. This data, being completely exogenous to the one used in currently active EWS, can offer a different and complementary perspective on the causes of such crises. We focus on the area of West Africa, which has been at the center of many humanitarian crisis since the beginning of this century. By exploiting state of the art text mining techniques on a corpus of textual documents in French (including video transcriptions extracted from the YouTube channels of four West African news broadcasters and news articles obtained from the online versions of two local newspapers of Burkina Faso) we will analyze food security situations in different regions of the study area in recent years, by also proposing a food security indicator based on textual data, namely T XT -FS.
“…For this, we compute by w2v the semantic similarity between each article and the generalist lexicon GLEX, used as a basis to identify articles on the theme "food security". The principle is to consider an article as dealing with food security if its semantic similarity with GLEX by w2v is higher than a threshold x (chosen and validated in the Appendix document [4]). This aims to detect the articles of interest to focus the analyses.…”
Section: Our Food Security Pipelinementioning
confidence: 99%
“…In this study, we examine the ability of text mining methods to extract and analyze the qualitative information used as proxies for the national and regional food situation and its evolution over the last ten years in Burkina Faso from a corpus of newspapers from the country. The aim is to provide explainable indicators complementary to the automatic predictions of food security scores, i.e., as the ones obtained in our previous works based on the application of machine learning approaches on heterogeneous [3] and textual [2] data.…”
Food security is a major concern in West Africa, particularly in Burkina Faso, which has been the epicenter of a humanitarian crisis since the beginning of this century. Early warning systems for food insecurity and famines rely mainly on numerical data for their analyses, whereas textual data, which are more complex to process, are rarely used. To this end, we propose an original and dedicated pipeline that combines different textual analysis approaches (e.g., word embedding, sentiment analysis, and discrimination calculation) to obtain an explanatory model evaluated on real-world and large-scale data. The results of our analyses have proven how our approach provides significant results that offer distinct and complementary qualitative information on the food security theme and its spatial and temporal characteristics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.