notesum.ai

Published at December 9

MISFEAT: Feature Selection for Subgroups with Systematic Missing Data

cs.LG

cs.DB

stat.ML

Released Date: December 9, 2024

Authors: Bar Genossar¹, Thinh On², Md. Mouinul Islam², Ben Eliav¹, Senjuti Basu Roy², Avigdor Gal¹

Aff.: ¹Technion – Israel Institute of Technology; ²New Jersey Institute of Technology

Arxiv: http://arxiv.org/pdf/2412.06711v1

Refer to caption

Patient ID	Age	Ethnicity	Blood Pressure	Family History	Body Weight	Smoking	Cholesterol	Readmission
1	45	Asian	Normal	Yes	Overweight	Non-smoker	Normal	No
2	62	Caucasian	Null	No	Overweight	Smoker	Normal	Yes
3	35	Caucasian	Normal	Yes	Normal	Non-smoker	Null	Yes
4	50	Caucasian	Null	Yes	Obese	Non-smoker	High	Yes
5	30	Asian	Null	No	Normal	Smoker	Null	No
6	28	Caucasian	Null	Yes	Overweight	Non-smoker	Null	Yes
7	55	Asian	Hypertension	Yes	Overweight	Smoker	Low	No
8	28	Asian	Null	No	Normal	Smoker	Null	No
9	60	Caucasian	Null	Yes	Obese	Non-smoker	Normal	Yes
10	38	Asian	Null	Yes	Overweight	Smoker	Null	Yes
11	54	Asian	Prehypertension	Yes	Normal	Non-smoker	Null	Yes
12	30	Caucasian	Prehypertension	No	Underweight	Smoker	Null	No