EWEH, FELICIA-MICHELLE MBIM (2026) WEB-BASED NUTRITION MONITORING AND FOOD RECOMMENDATION SYSTEM FOR CHILDREN OF 1-5 YEARS IN ENUGU SATE (GrowWell). Other thesis, Godfrey Okoye University, Enugu.
|
Text
EWEH FELICIA-MICHELLE MBIM project.pdf Download (1MB) | Preview |
Abstract
Child malnutrition remains a significant public health challenge among children aged 1 to 5 years in Enugu State, Nigeria, with many children suffering from stunting, wasting, and underweight conditions due to inadequate nutritional monitoring and poor dietary practices at the household level. Existing digital platforms such as RapidPro were designed for health workers and programme managers rather than caregivers, and none incorporated machine learning-based nutritional classification, locally adapted food recommendations, or structured feeding timetables using locally available Enugu State foods. This study designed and implemented GrowWell, a web-based Nutrition Monitoring and Food Recommendation System for children aged 1 to 5 years in Enugu State. The system was developed using the Object Oriented Analysis and Design (OOAD) methodology and implemented using Python, Flask, MySQL, HTML, CSS, and JavaScript. The K-Nearest Neighbors (KNN) machine learning algorithm was trained on a synthetic dataset of 1,000 child records generated using WHO Child Growth Standards (2006) to classify children's nutritional status as Normal, Underweight, Stunted, Wasted, or Severely Wasted based on Weight-for-Age (WAZ), Height-for-Age (HAZ), and Weight-for-Height (WHZ) z-scores. The system generates personalized food recommendations using twenty-eight locally available Enugu State foods and produces a structured 2-week feeding timetable tailored to each child's classified nutritional status and age group. The KNN model achieved an overall classification accuracy above 90% on the test dataset, and all fifteen functional test cases passed successfully. User acceptance testing confirmed that the system is user-friendly and locally relevant. The GrowWell system demonstrates the potential of web-based machine learning systems to improve child nutrition monitoring and dietary guidance at the household level in Enugu State, Nigeria.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science |
| Depositing User: | MICHAEL MADUBUKO |
| Date Deposited: | 22 Jul 2026 13:32 |
| Last Modified: | 22 Jul 2026 13:32 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/6028 |
Actions (login required)
 |
View Item |
