Clustering of the West African Starchy Roots and Tubers using Nutritive Value

Abstract

The objective of this study was to sub-group the  West African starchy roots and tubers’ food  sources according to their nutritive values. The k means algorithm was deployed on a dataset  consisting of starchy roots and tubers food items  extracted from the West African Food  Composition Table. Various measures of evaluating clustering validity were employed and  the results showed that the clustering was valid.  Three clusters were extracted, each consisting of  food sources that are very similar in nutritive  value. Findings prove that in terms of nutritive  value, some kinds of yam and cassava could be  classified together, while some kinds of sweet  potato and cocoyam could also be classified  together, and so on. The clusters can be explored  by nutritionists, dieticians, food processing  enterprises, and food scientists to find alternative  food sources when their original choices are  unavailable. Though clustering validation  showed that food sources within the same cluster  are significantly similar on a general note, a future  study is required to research on the extent to which  food sources within the same cluster are similar to  each other at a granular level. 

https://doi.org/10.59200/ICONIC.2022.025