We have conducted a study in Northern Bangladesh on Hygiene practice at household level. Hygiene practices were checked with yes/no responses in different dimensions. We want to see regional variations in hygiene practice and its association with education of the respondent. Can we use cluster analysis?
Dear Md. Ektear Uddin
These links might be useful, have a look:
1. https://stats.stackexchange.com/questions/40613/why-dont-dummy-variables-have-the-continuous-adjacent-category-problem-in-clust
2. https://datascience.stackexchange.com/questions/22/k-means-clustering-for-mixed-numeric-and-categorical-data
3. https://www.ibm.com/support/pages/clustering-binary-data-k-means-should-be-avoided
Kind Regards
Qamar Ul Islam
There are some problem where we cannot apply ML algorithms. Similarly, it is inappropriate to apply KNN like clustering algorithm on 'Yes/No' nominal data. I feel the direct application of clustering algorithm on this data will not give results. If anything relevant, kindly update. But, your logic is correct.
Hi. I hope the following link might help:
https://towardsdatascience.com/clustering-datasets-having-both-numerical-and-categorical-variables-ed91cdca0677?gi=192501566055
Can. To do this, you first need to metrize the nominal scales.
https://www.researchgate.net/publication/331801337
For example, as in the Eidos system: http://lc.kubagro.ru/aidos/The_Eidos_en.htm
Check https://www.researchgate.net/publication/228758935_Cluster_analysis_and_categorical_data
Also check https://www.statisticssolutions.com/free-resources/directory-of-statistical-analyses/cluster-analysis/
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Cell Biology
- Culture Cells
- Cell Line