TheDispensersforSafeWaterprogramunderEvidenceActionpromotespoint of collection water treatment through the installation of chlorine dispenser gadgets in rural parts of Kenya. Although the initiative has improved access to safe drinking water, monitoring household adoption remained a challenge during the COVID-19 pandemic, which limited field-based data collection and led to increased dependence on phone surveys. In addition, technology adoption data are often imbalanced, which poses difficulties for traditional classification methods. This study aimed to develop and implement a stacking ensemble classifier to model the adoption of chlorine dispensers among households in western Kenya. Data were collected from 27,457 households. The analysis used structured household, promoter and spot check survey data. The key variables included chlorine availability, user knowledge, household demographics, and engagement with promoters. RF, ANN, and NB models were trained and evaluated individually, then combined using a stacked ensemble approach. The ensemble model outperformed all base learners, achieving the highest accuracy (69.1%) and AUC (0.6959). The variable importance analysis revealed that the presence of chlorine and the knowledge of the user were the strongest predictors of adoption. In conclusion, ensemble learning provides a reliable method for modeling behavioral adoption in public health interventions. The findings offer practical insights for programs and demonstrate the potential of machine learning in improving, targeting and monitoring of safe water initiatives in low-resource settings.
| Published in | International Journal of Data Science and Analysis (Volume 11, Issue 6) |
| DOI | 10.11648/j.ijdsa.20251106.12 |
| Page(s) | 171-177 |
| Creative Commons |
This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited. |
| Copyright |
Copyright © The Author(s), 2025. Published by Science Publishing Group |
Stacked Ensemble, Random Forest, Artificial Neural Networks, Naive Bayes, Total Chlorine Residue, Machine Learning
| [1] | Millennium Water Alliance. Learning from Piloting Dispensers for Safe Water. In: Washington, DC (2019). |
| [2] | Meriem Amrane et al. Breast cancer classification using machine learning. In: 2018 Electric Electronics, Computer Science, Biomedical Engineering Meeting (EBBT) (2018), pp. 1-4. |
| [3] | Thomas G. Dietterich. Ensemble Methods in Machine Learning. In: International Workshop on Multiple Classifier Systems. Berlin, Heidelberg: Springer Berlin Heidelberg, June 2000, pp. 1-15. |
| [4] | M. Husein et al. Examination of water quality at the household level and its association with diarrhea among young children in Ghana: Results from UNICEF-MICS6 survey. In: PLOS Water (2023). |
| [5] | Y. Iijima et al. High Prevalence of Diarrheagenic Escherichia coli among Children with Diarrhea in Kenya. In: Japanese Journal of Infectious Diseases 70.1 (2017), pp. 80-83. |
| [6] | H. Jallow et al. Transfer learning for predicting gross domestic product growth based on remittance inflows using RNN-LSTM hybrid model: a case study of The Gambia. In: Frontiers in Artificial Intelligence 8 (2025), p. 1510341. |
| [7] | Jinseok Kim and Jenna Kim. The impact of imbalanced training data on machine learning for author name disambiguation. In: Scientometrics 117.1 (2018), pp. 511-526. |
| [8] | Sammy Kiprop et al. Classification of Contraceptive Use Among Undergraduate Students Using a Supervised Machine Learning Technique. In: American Journal of Theoretical and Applied Statistics 12.5 (2023), pp. 120 128. |
| [9] | Dian Kurniawati et al. Poor Basic Sanitation Impact on Diarrhea Cases in Toddlers. In: Journal Name 13 (2021), pp. 41-47. |
| [10] | Daniele Lantagne. Sodium hypochlorite dosage for household and emergency water treatment. In: Journal American Water Works Association 100 (Aug. 2008), pp. 106119. |
| [11] | S. Mooney and V. Pejaver. Big Data in Public Health: Terminology, Machine Learning, and Privacy. In: Annual Review of Public Health 39 (2018), pp. 95-112. |
| [12] | K. J. Nath, Sally Bloomfield, and Martin Jones. Household water storage, handling and point-of-use treatment. In: 2006. |
| [13] | Matilda Q. R. Pembury Smith and Graeme D. Ruxton. Effective use of the McNemar test. In: Behavioral Ecology and Sociobiology 74.11 (2020), p. 133. |
| [14] | Hasan Ahmed Salman, Ali Kalakech, and Amani Steiti. Random Forest Algorithm Overview. In: Babylonian Journal of Machine Learning 2024 (2024), pp. 69-79. |
| [15] | K. Ásia Chaves da Silva and C. N. K. Suda. Acute diarrheal diseases and their relationship with water quality in Araguatins, Tocantins, Brazil: a cross-sectional study. In: Bioscience Journal (2023). |
| [16] | Bayu AdhiTama, SunIm, andSeungchulLee. Improving an Intelligent Detection System for Coronary Heart Disease Using a Two-Tier Classifier Ensemble. In: BioMed Research International (2020). |
| [17] | Yu-chen Wu and Jun-wen Feng. Development and Application of Artificial Neural Network. In: Wireless Personal Communications 102.2 (2018), pp. 1645-1656. |
| [18] | Feng-Jen Yang. An Implementation of Naive Bayes Classifier. In: 2018 International Conference on Computational Science and Computational Intelligence (CSCI). 2018, pp. 301-306. |
APA Style
Oluoch, R. O., Imboga, H., Waititu, A., Mwelu, S., Evance, I., et al. (2025). Stacked Ensemble Classifier for Adoption of Point-of-Collection Water Treatment Technology Among Households in Western Kenya. International Journal of Data Science and Analysis, 11(6), 171-177. https://doi.org/10.11648/j.ijdsa.20251106.12
ACS Style
Oluoch, R. O.; Imboga, H.; Waititu, A.; Mwelu, S.; Evance, I., et al. Stacked Ensemble Classifier for Adoption of Point-of-Collection Water Treatment Technology Among Households in Western Kenya. Int. J. Data Sci. Anal. 2025, 11(6), 171-177. doi: 10.11648/j.ijdsa.20251106.12
Share This Article
Twitter
Linked In
Facebook
Copy
Download@article{10.11648/j.ijdsa.20251106.12,
author = {Robert Omondi Oluoch and Herbert Imboga and Anthony Waititu and Susan Mwelu and Illah Evance and Ferdnand Ongeta},
title = {Stacked Ensemble Classifier for Adoption of Point-of-Collection Water Treatment Technology Among Households in Western Kenya
},
journal = {International Journal of Data Science and Analysis},
volume = {11},
number = {6},
pages = {171-177},
doi = {10.11648/j.ijdsa.20251106.12},
url = {https://doi.org/10.11648/j.ijdsa.20251106.12},
eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijdsa.20251106.12},
abstract = {TheDispensersforSafeWaterprogramunderEvidenceActionpromotespoint of collection water treatment through the installation of chlorine dispenser gadgets in rural parts of Kenya. Although the initiative has improved access to safe drinking water, monitoring household adoption remained a challenge during the COVID-19 pandemic, which limited field-based data collection and led to increased dependence on phone surveys. In addition, technology adoption data are often imbalanced, which poses difficulties for traditional classification methods. This study aimed to develop and implement a stacking ensemble classifier to model the adoption of chlorine dispensers among households in western Kenya. Data were collected from 27,457 households. The analysis used structured household, promoter and spot check survey data. The key variables included chlorine availability, user knowledge, household demographics, and engagement with promoters. RF, ANN, and NB models were trained and evaluated individually, then combined using a stacked ensemble approach. The ensemble model outperformed all base learners, achieving the highest accuracy (69.1%) and AUC (0.6959). The variable importance analysis revealed that the presence of chlorine and the knowledge of the user were the strongest predictors of adoption. In conclusion, ensemble learning provides a reliable method for modeling behavioral adoption in public health interventions. The findings offer practical insights for programs and demonstrate the potential of machine learning in improving, targeting and monitoring of safe water initiatives in low-resource settings.
},
year = {2025}
}
TY - JOUR T1 - Stacked Ensemble Classifier for Adoption of Point-of-Collection Water Treatment Technology Among Households in Western Kenya AU - Robert Omondi Oluoch AU - Herbert Imboga AU - Anthony Waititu AU - Susan Mwelu AU - Illah Evance AU - Ferdnand Ongeta Y1 - 2025/11/10 PY - 2025 N1 - https://doi.org/10.11648/j.ijdsa.20251106.12 DO - 10.11648/j.ijdsa.20251106.12 T2 - International Journal of Data Science and Analysis JF - International Journal of Data Science and Analysis JO - International Journal of Data Science and Analysis SP - 171 EP - 177 PB - Science Publishing Group SN - 2575-1891 UR - https://doi.org/10.11648/j.ijdsa.20251106.12 AB - TheDispensersforSafeWaterprogramunderEvidenceActionpromotespoint of collection water treatment through the installation of chlorine dispenser gadgets in rural parts of Kenya. Although the initiative has improved access to safe drinking water, monitoring household adoption remained a challenge during the COVID-19 pandemic, which limited field-based data collection and led to increased dependence on phone surveys. In addition, technology adoption data are often imbalanced, which poses difficulties for traditional classification methods. This study aimed to develop and implement a stacking ensemble classifier to model the adoption of chlorine dispensers among households in western Kenya. Data were collected from 27,457 households. The analysis used structured household, promoter and spot check survey data. The key variables included chlorine availability, user knowledge, household demographics, and engagement with promoters. RF, ANN, and NB models were trained and evaluated individually, then combined using a stacked ensemble approach. The ensemble model outperformed all base learners, achieving the highest accuracy (69.1%) and AUC (0.6959). The variable importance analysis revealed that the presence of chlorine and the knowledge of the user were the strongest predictors of adoption. In conclusion, ensemble learning provides a reliable method for modeling behavioral adoption in public health interventions. The findings offer practical insights for programs and demonstrate the potential of machine learning in improving, targeting and monitoring of safe water initiatives in low-resource settings. VL - 11 IS - 6 ER -
Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya
Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya
Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya
Department of Statistics and Actuarial Science, Jomo Kenyatta University of Agriculture and Technology, Juja, Kenya
Evidence Action, Africa Regional Office, Nairobi, Kenya
Evidence Action, Africa Regional Office, Nairobi, Kenya
Information