Health Risk Profiles of 800 Older Adults in Kerala via Machine Learning Clustering

Kerala, India, is the geographic scope for this dataset of 800 community-dwelling adults aged 60 years and above. The data was analyzed by Blessy Sarah Mathew using an integrated machine learning framework combining K-Means clustering, PCA, and Random Forest classification to identify multidimensional health risk profiles. The dataset was last updated on 2026-04-17.

Use Cases

• Clustering older adults into risk profiles based on digital health literacy, functional difficulty, and self-management capacity.
• Training supervised classifiers to predict cluster membership using demographic and socioeconomic features.
• Analyzing associations between socioeconomic factors like education and income with health risk profiles.
• Visualizing multidimensional cluster separation using dimensionality reduction techniques like PCA.

Strengths

• Dataset size is 558.9 KB, indicating a focused and manageable scope for analysis.
• Analysis framework is detailed, including K-Means clustering, PCA, and Random Forest classification.
• Clustering achieved an average Silhouette score of 0.30, indicating some internal validity.
• Data covers 800 individuals, providing a basis for statistical testing of between-cluster differences.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data is from a single cross-sectional survey in Kerala, which may limit generalizability to other regions.

Provenance

Source

Author: Blessy Sarah Mathew. Platform: figshare.

Collection Method

Cross-sectional survey of 800 community-dwelling older adults.

Freshness

Last updated 2026-04-17 10:26:36; freshness should be verified.

Geography

Kerala, India.