Towards Harnessing Phone Messages and Telephone Conversations for Prediction of Food Crisis

Abstract

Food insecurity is a global challenge affecting millions of people especially those from least developed regions. Famine predictions are being carried out to estimate when shortage of food is most likely to happen. The traditional data sets such as house hold information, price trends, crop production trends and biophysical data used for predicting food insecurity are both labor intensive and expensive to acquire. Current trends are towards harnessing big data to study various phenomena such sentiment analysis and stock markets. Big data is said to be easier to obtain than traditional datasets. This study shows that phone messages archives and telephone conversations as big datasets are potential for predicting food crisis. This is timely with the current situation of massive penetration of mobile technology and the necessary data can be gathered to foster studies such as this. Computation techniques such as Naïve Bayes, Artificial Networks and Support Vector Machines are prospective candidates in this strategy. If the strategy is to work in a nation like Uganda, areas of concern have been highlighted. Future work points at exploring this approach experimentally.

Full text

International Journal of System Dynamics Applications, 4(4), 1-16, October-December 2015 1
Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Keywords: Big Data, Food Insecurity, Machine Learning Techniques, Phone Messages, Telephone
Conversations,TextExtractionMethods
ABSTRACT
Foodinsecurityisaglobalchallengeaffectingmillionsofpeopleespeciallythosefromleastdevelopedregions.
Faminepredictionsarebeingcarriedouttoestimatewhenshortageoffoodismostlikelytohappen.The
traditionaldatasetssuchashouseholdinformation,pricetrends,cropproductiontrendsandbiophysical
datausedforpredictingfoodinsecurityarebothlaborintensiveandexpensivetoacquire.Currenttrendsare
towardsharnessingbigdatatostudyvariousphenomenasuchsentimentanalysisandstockmarkets.Bigdata
issaidtobeeasiertoobtainthantraditionaldatasets.Thisstudyshowsthatphonemessagesarchivesand
telephoneconversationsasbigdatasetsarepotentialforpredictingfoodcrisis.Thisistimelywiththecurrent
situationofmassivepenetrationofmobiletechnologyandthenecessarydatacanbegatheredtofosterstudies
suchasthis.ComputationtechniquessuchasNaïveBayes,ArticialNetworksandSupportVectorMachines
areprospectivecandidatesinthisstrategy.IfthestrategyistoworkinanationlikeUganda,areasofconcern
havebeenhighlighted.Futureworkpointsatexploringthisapproachexperimentally.
Towards Harnessing
Phone Messages and
Telephone Conversations for
Prediction of Food Crisis
AndrewLukyamuzi,InstituteofComputerScience,MbararaUniversityofScienceand
Technology,Mbarara,Uganda
JohnNgubiri,CollegeofComputingandInformationSciences,MakerereUniversity,
WashingtonOkori,UgandaTechnologyandManagementUniversity(UTAMU),Kampala,
Uganda
1. INTRODUCTION
Food insecurity is among of the terrors that have perturbed human welfare. Humans have lived
with this challenge for many generations. Numerous sources have been documented to testify
early manifestations of this challenge. Such sources include; scholarly works, religious books,
story books and oral traditional. The famines that struck Europe in the years between 1343 and
1345 were deadly (Mellor, 1987). About 43 million people lost their lives in these famines. In
DOI: 10.4018/IJSDA.2015100101

Copyright © 2015, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
2 International Journal of System Dynamics Applications, 4(4), 1-16, October-December 2015
the period from 1959 to 1961China fell into a similar trap (Mellor, 1987). It is estimated that
between 16 and 64 million people in China perished for the same cause.
The effects of food insecurity especially those that were devastating compelled the world
to seek appropriate solutions. As a result, success stories of reduced food insecurity cases have
been recorded. According to Food and Agricultural Organization et al. (2014) hunger cases have
reduced by 100 million people in the previous decade. The Millennium Development Goal one
had a target to reduce hunger cases by half not later than 2015. Among the countries that have
achieved this target, Latin America and the Caribbean have made the greatest progress (Food
and Agricultural Organization et al., 2014).
It is important to examine the extent at which the world has managed to control or eliminate
food insecurity. Unfortunately the hard fact remains that food insecurity is still a big challenge.
The available statistics give limited room to doubt this. About 842 million populations in world
are victims of chronic hunger (Food and Agricultural Organization et al, 2013). In the report
released by Food and Agricultural Organization et al. (2014) it was established that 805 million
populations are chronically undernourished. The Committee on World Food Security (2013) has
disclosed that more than 200 million children under five years of age are malnourished. In the
period from 1995-98 about 1 million people lost their lives in the famine attacks of North Korea
(Committee for Human Rights in North Korea, 2005). In order to keep pace with population
increase by the year 2050, food production should increase by 70% (International Fund for Agri-
cultural Development, 2010). On the other hand factors such as climate change, soil exhaustion,
and bio fuel practices are exacerbating this challenge (Faaij, 2008). It is therefore evident that
any innovation that can assist in transmuting this challenge is worthwhile.
Prediction of food insecurity is a possible remedy to this challenge. This is instrumental
in guiding stakeholders where to direct early intervention reliefs. These reliefs are helpful in
several ways: (1) the impact of food insecurity can be reduced or eliminated completely, and
(2) the expense involved in amelioration can be minimized (Okori and Obua, 2011; Brown et
al, 2008). This has proved successful in some parts of the world. For instance, according to the
United States Department of Agriculture –USDA (2005), reliable monitoring of food insecurity
contributes to the effective operation of Federal programs, food assistance programs, and other
government initiatives aimed at reducing food insecurity.
Several features have been proposed for prediction of food insecurity. These include but
not limited to house hold information, price trends, biophysical features, and economic growth.
Attention is needed in determining features suitable for predicting of food insecurity. While
features used for one study can be applied to other studies, this is not always appropriate. This
is because players for food dynamics are not necessarily the same. There situations when these
features are the same but portray variation in the relevance. This can have a big impact on predic-
tion performance. Inspection is a possible approach in determining which features are appropri-
ate for a particular study. This involves examining the role of various features in food security
dynamics. Correlations are commonly done to establish the strength between food insecurity
and the proposed features. In this way the research is able to establish which features are appro-
priate for prediction of food insecurity. The second method is to review relevant literature. By
examining previous related studies, it is possible to deduce justifications for selection of certain
features as prediction parameters. In the process, patterns governing choice of the features and
the environment can be established. These patterns now become an important reference point
in determining which predictive features to use for a given study.
Identifying appropriate predictive features for use in a given study is not enough. It is equally
important to consider the resources and time the required to gather the required the datasets. Data
sets for features such as household information, biophysical properties, and price trends require

14 more pages are available in the full version of this
document, which may be purchased using the "Add to Cart"
button on the product's webpage:
www.igi-global.com/article/towards-harnessing-phone-
messages-and-telephone-conversations-for-prediction-of-
food-crisis/136995?camid=4v1
This title is available in InfoSci-Journals, InfoSci-Journal
Disciplines Computer Science, Security, and Information
Technology, InfoSci-Computer Systems and Software
Engineering eJournal Collection, InfoSci-Select. Recommend
this product to your librarian:
www.igi-global.com/e-resources/library-
recommendation/?id=2
Related Content
Model of Meta-Reflection and In-Depth Analysis of Knowledge Creation
Process in Design
Georgi V. Georgiev and Yukari Nagai (2011). International Journal of Knowledge and
Systems Science (pp. 49-67).
www.igi-global.com/article/model-meta-reflection-depth-
analysis/54737?camid=4v1a
Decision Support Fundamentals
(2013). Decision Control, Management, and Support in Adaptive and Complex
Systems: Quantitative Models (pp. 1-44).
www.igi-global.com/chapter/decision-support-
fundamentals/74432?camid=4v1a

Introduction to Control Systems Design Using Matlab
Lazaros Moysis, Ahmad Taher Azar, Ioannis Kafetzis, Michail Tsiaousis and Nikolaos
Charalampidis (2017). International Journal of System Dynamics Applications (pp.
130-170).
www.igi-global.com/article/introduction-to-control-systems-design-using-
matlab/182822?camid=4v1a
Beernet: Building Self-Managing Decentralized Systems with Replicated
Transactional Storage
B. Mejías and P. Van Roy (2012). Technological Innovations in Adaptive and
Dependable Systems: Advancing Models and Concepts (pp. 282-304).
www.igi-global.com/chapter/beernet-building-self-managing-
decentralized/63587?camid=4v1a