Implementing K-Means clustering for research about environmental awareness and environmental practices of Ecuadorian households
This research describes the procedure carried out for the generation and analysis of clusters of Ecuadorian households according to their practices and environmental awareness. After a selection of variables of interest and a treatment of the data set, the K-means algorithm is used, supported by techniques such as the Elbow Method to group the Ecuadorian households surveyed.
The results are discussed with an approach that seeks to classify Ecuadorian households according to their level of environmental awareness and environmental practices.
The pdf file "Clustering_ambiente.pdf" is the paper that explains in more detail all the procedure and results, a translation of this paper to english is pending.
In this repository we show all the implemented python scripts, images shown in the final paper and more that for brevity sake were not added to the paper. This images complement the results of our findings.
Some steps (mostly of data extraction) were made in Weka, this is explained in the paper.
It is important to say this was an independent research, that without doubt deserves a review of methods and steps taken. Nevertheless, a low of work was put into this research, thus obtaining good results.
I. N. de E. y Censos, “Hogares”, Instituto Nacional de Estadística y Censos. https://www.ecuadorencifras.gob.ec/hogares/ (acessed feb. 22, 2021).