Participant Characteristics for Smoking Cessation JITAI Algorithm Optimization

Thirty-seven participants completed 16 ecological momentary assessments per day for the first 10 days of a smoking cessation attempt, reporting mood, context, behavior, cravings, and smoking lapses. The dataset, authored by Corinna Leppin and shared on figshare under CC-BY-4.0, contains participant-level characteristics used to evaluate random forest algorithms predicting lapse and craving risk. Performance metrics for these predictions, including F1-scores and ROC-AUC, are reported with substantial inter-individual variability.

Use Cases

• Optimizing ecological momentary assessment (EMA) prompt frequency based on reported lapse and craving prediction performance.
• Evaluating feature reduction strategies for machine learning models in behavioral health interventions.
• Comparing algorithm performance when trained with participant-specific data versus general data.
• Analyzing the trade-off between participant burden (EMA frequency) and predictive model accuracy.

Strengths

• Includes detailed performance metrics (F1-score, ROC-AUC) for lapse and craving prediction across different experimental configurations.
• Based on data from 37 participants completing a high-frequency (16 per day) EMA protocol over 10 days.
• Explicitly tests the impact of varying EMA frequency, predictor count, and training data source on algorithm performance.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 9.5 KB, indicating limited scope and likely summary-level data.

Provenance

Source

Corinna Leppin

Collection Method

Ecological momentary assessments (EMAs) collected from participants during a smoking cessation attempt.

Freshness

Last updated 2026-05-14 17:33:12; freshness should be verified.