BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:HAI & SDS Seminar with Dan Iancu and Antonio Skillicorn
X-WR-TIMEZONE:Pacific Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260513T164150Z
UID:tag:localist.com\,2008:EventInstance_51570064994945
DTSTART:20260318T190000Z
DTEND:20260318T201500Z
DESCRIPTION:Interpretable Machine Learning and Mixed Datasets for Predictin
 g Child Labor in Ghana’s Cocoa Sector\n\nChild labor remains prevalent i
 n Ghana’s cocoa sector and is associated with adverse educational and he
 alth outcomes for children. This exploratory work examines how two surveys
  that measure child labor in Ghana (NORC and GLSS7)\, but differ in qualit
 y and scale\, can be jointly leveraged for less biased prediction and to i
 dentify key predictors of child labor risk. We further investigate whether
  district-level satellite indicators\, including yield-weighted cocoa-driv
 en deforestation\, newly lit area\, and newly urban area\, enhance predict
 ive performance and play important roles in shaping model predictions. Usi
 ng non-parametric machine learning models (XGBoost\, Random Forest) paired
  with cross-validation and a hyperparameter grid search\, we find that the
  best-performing model in classifying child laborers achieves an out of sa
 mple AUC of 0.95 and F1 of 0.84. Model interpretability tools (SHAP values
 \, partial dependence plots) highlight influential predictors such as chil
 d age\, cocoa-driven deforestation\, school commute time\, newly lit area\
 , and household herbicide expenditures. In addition to emerging as the sec
 ond most explanatory feature\, cocoa-driven deforestation also shows a cle
 ar nonlinear association with predicted child labor risk. Our approach dem
 onstrates new ways of grappling with data scarcity and bias in child labor
  measurement\, while our findings provide actionable risk profiles to supp
 ort monitoring efforts and underscore the complex interconnections between
  child labor and environmental practices.
GEO:37.429987;-122.17333
LOCATION:Gates Computer Science Building\, 119
SUMMARY:HAI & SDS Seminar with Dan Iancu and Antonio Skillicorn
URL;VALUE=URI:https://events.stanford.edu/event/hai-seminar-with-dan-iancu-
 sarah-billington-antonio-skillicorn
CATEGORIES:Class/Seminar
END:VEVENT
END:VCALENDAR