| Assessing Anemia Prevalence Among Women: A Comprehensive Analysis Using Machine Learning Approaches with the NFHS-5 Dataset |
| Authors: |
Manjusha Tippavajhula, P. Pranay, A. Rajini & Raju Kommarajula |
| Source: |
Springer Nature, Volume 117 |
| Topic(s): |
Anemia
Data models
Modelling
Women's health
|
| Country: |
Asia
India
|
| Published: |
OCT 2025 |
| Abstract: |
Anemia, characterized by low hemoglobin levels, poses a substantial health risk among women aged 15-49 years. This research employs multinomial logistic regression to explore the correlation between independent variables and distinct categorical levels of anemia. The study aims to refine the model iteratively while identifying significant predictors influencing anemia levels. The analysis demonstrates an overall model accuracy of 77.9%, indicating the model’s ability to classify individuals into their respective anemia levels. Notably, it excels at predicting moderate anemia (83.1%). However, the model encounters challenges in accurately predicting severe and mild anemia, achieving 0% accuracy in the latter category. Predictions for non-anemic cases are also for targeted interventions and further research. However, it is crucial to consider the strengths and limitations of the model when implementing effective public health strategies. Using the NFHS-5 dataset in a novel way and sophisticated multinomial logistic regression algorithms, this work offers insightful information for focused public health interventions and policy suggestions. Our findings greatly add to the body of knowledge already available on women’s health by providing a thorough grasp of the anemia predictors. |
| Web: |
https://doi.org/10.1007/978-981-96-4410-0_6 |
|