Supervised Learning Performance Metrics for Pediatric Diarrhea Prediction in Sierra Leone

2020 data from the Sierra Leone Demographic and Health Survey (SLDHS) provides performance metrics for machine learning models predicting diarrhea in children under five. The dataset, authored by Yahye Hassan Muse, contains model evaluation results in a 9.5 KB Excel file.

Use Cases

• Compare model performance metrics like accuracy and F1-score to identify the best algorithm for pediatric diarrhea prediction.
• Analyze the relationship between reported performance metrics and the specific prediction task for diarrhea in under-five children.
• Validate new predictive models against the benchmark metrics established in this SLDHS 2020 analysis.

Strengths

• Metrics are derived from the nationally representative Sierra Leone Demographic and Health Survey (SLDHS) 2020.
• The dataset is focused on a specific public health prediction task for a vulnerable demographic (children under five).
• Data is openly licensed under CC BY 4.0 for reuse and redistribution.

Limitations

• The dataset is very small at 9.5 KB, indicating it contains summary results, not the underlying training or patient-level data.
• Without access to the raw columns or sample data, the specific performance metrics (e.g., precision, recall) and model names are unknown.
• The data is static from a 2020 survey and may not reflect current model performance or disease patterns.

Provenance

Source

figshare, authored by Yahye Hassan Muse.

Collection Method

Derived from analysis of the Sierra Leone Demographic and Health Survey (SLDHS) 2020.

Time Range

2020.

Freshness

null

Geography

Sierra Leone.