A non$-$governmental organisation $($NGO$)$ is applying for a grant to support them in

combating child malnutrition across several countries worldwide. To increase their

chances of securing funding, they have decided to analyse reputable databases to

find relevant information that can support their campaign for improved child

nutrition across the globe. They have found some datasets by the World Health

Organisation $($WHO$)$ which seem to suit their needs. These datasets contain statistics

on infant breastfeeding rates as well as child mortality rates $(0-5$ years$)$ across

different countries and year periods. Now, they have to present a strong argument

for the correlation between these datasets and how this relates to their child

nutrition campaign.

You have been consulted as an AI engineer to help with the analysis of these

datasets and to reveal insights from them as follows:

$$ Your initial observations from your exploration of these datasets are as

follows:

$$ The data contains some missing values

$$ The data mostly contains numeric values which may not be properly

formatted

$$ The data contains relatively few features

$$ Both datasets seem to cover different ranges of years.

In your meeting with your client, you have agreed to use relevant Machine Learning

$($ML$)$ techniques $($supervised$,$ unsupervised, etc.$)$ and AI search or optimisation

techniques. You are expected to present a report to your client by constructing a

robust model which must follow the guidelines presented below:

$1.$ Preprocess the datasets to create a single dataset which contains the needed

information to predict mortality rates for different years for each country.

$2.$ Use AI search or optimisation techniques $($whichever is appropriate$)$ to align

the year periods for each country across all the datasets.

$3.$ Based on the dataset you have created, build a supervised or unsupervised

ML model to predict the mortality rates for each country for the different

years possible.

$4.$ Justify your design decisions for tasks $1,2$ & $3.$

$5.$ Critically evaluate the learning model you have built

$6.$ Evaluate the robustness of your model by applying appropriate validation

techniques $($and identifying a suitable subset of data for validation$).$

While setting the parameters of the search or optimisation method, pay special

attention to selecting appropriate metrics $($evaluation criteria$).$ The chosen metrics

$2$

will play a critical role in the relative success or failure of the potential solution$($s$)$ and

in setting the direction of the search or optimisation.