Smoking Cessation JITAI Study Data on Lapse and Craving Prediction

Thirty-seven participants completed 16 ecological momentary assessments per day for 10 days, reporting mood, context, behavior, cravings, and smoking lapses. The dataset contains results from random forest algorithms predicting lapses and cravings, evaluated via F1-score and ROC-AUC. The data was authored by Corinna Leppin and last updated on 2026-05-14.

Use Cases

• Evaluating machine learning model performance for predicting smoking lapses based on EMA-reported mood, context, and behavior.
• Analyzing the trade-off between EMA prompt frequency and prediction accuracy for lapse and craving risk.
• Investigating the impact of feature reduction and participant-specific training data on algorithm performance in a JITAI context.

Strengths

• Contains performance metrics (F1-score, ROC-AUC) for lapse and craving prediction from a study with 37 participants.
• Includes results from systematically varied experimental conditions, such as EMA frequency (3 to 16 prompts/day) and predictor count.
• Data is shared under a CC-BY-4.0 license, allowing for broad reuse.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 5.5 KB, indicating limited scope, likely containing summary results rather than raw participant data.

Provenance

Source

figshare

Collection Method

Data likely originates from a research study where participants completed ecological momentary assessments (EMAs) during a smoking cessation attempt.

Freshness

Last updated 2026-05-14 17:33:13; freshness should be verified.