BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Oxygen abundances as a key predictor of small planet hosts using g
 radient boosted decision trees
DTSTART;VALUE=DATE-TIME:20261028T165500Z
DTEND;VALUE=DATE-TIME:20261028T170500Z
DTSTAMP;VALUE=DATE-TIME:20261025T203357Z
UID:indico-contribution-3635@nyx.physics.mcgill.ca
DESCRIPTION:Speakers: Rosalind Liang (University of Victoria)\nUsing a nov
 el crossmatch of GALAH DR4 stellar abundances with K2 and TESS planet dete
 ctions\, we apply the gradient boosted decision tree ensemble model XGBoos
 t to investigate relative importances of specific chemical abundances and 
 molar ratios in predicting small planet hosts. In contrast to previous stu
 dies\, we find a consistently high feature importance for oxygen across mo
 dels. Using a probability threshold of P > 0.80\, the [X/O] ensemble ident
 ifies 1\,218 likely small-planet hosts\, including 13 of 133 K2/TESS candi
 dates. Abundance distributions reveal that the observed signal may be driv
 en primarily by oxygen depletion. These results provide a targeted shortli
 st for follow-up observations while highlighting the need for larger host 
 samples and more granular comparisons between planetary subclasses.\n\nhtt
 ps://nyx.physics.mcgill.ca/event/587/contributions/3635/
LOCATION:Online 
URL:https://nyx.physics.mcgill.ca/event/587/contributions/3635/
END:VEVENT
END:VCALENDAR