Research Article

Feature Signature Discovery for Autism Detection: An Automated Machine Learning Based Feature Ranking Framework

Table 2

Kaggle toddler dataset description.

FeatureDomainDescription

A1: response_to_nameBinary (0, 1)Does your child look at you when you call his/her name?
A2: eye_contactBinary (0, 1)How easy is it for you to get eye contact with your child?
A3: point_to_objectsBinary (0, 1)Does your child point to indicate that s/he wants something?
A4: sharing_interestBinary (0, 1)Does your child point to share an interest with you?
A5: pretend_playBinary (0, 1)Does your child pretend?
A6: follow_lookingBinary (0, 1)Does your child follow where you are looking?
A7: comfort_someoneBinary (0, 1)Does your child show signs of wanting to comfort someone upset?
A8: first_wordsBinary (0, 1)Description of child's first words
A9: simple_gestureBinary (0, 1)Does your child use simple gestures?
A10: stare_at_nothingBinary (0, 1)Does your child stare at nothing with no apparent purpose?
AgeNumericToddlers (months) and others (years)
Score by Q-chat-10Numeric1–10 (less than or equal to 3—no ASD traits; >3 ASD traits
SexM/FMale or female
EthnicityStringList of common ethnicities in text format
Born with jaundiceBoolean (Y/N)Whether the case was born with jaundice
Family member with ASD historyBoolean (Y/N)Whether any immediate family member has a PDD
Who is completing the testStringParent, self, caregiver, medical staff, clinician, etc
Class variable (ASD)/targetStringASD traits or No ASD traits (Yes/No)