Browse Teach Curate

Nature-Inspired Feature Selection for Autism Detection from Gait Video

A classification pipeline that detects Autism Spectrum Disorder from 3D walking-video gait data by selecting discriminative features with nature-inspired metaheuristic optimizers (chiefly a gravitational search algorithm) before supervised classification. A ranking-coefficient rule for choosing the initial leading particle reduces computation time, and a random-forest classifier with gravitational-search feature selection achieves the highest reported accuracy.

Rosie Text

Empirical Analysis of Nature-Inspired Algorithms for Autism Spectrum Disorder Detection Using 3D

Primary machine-learning research on detecting Autism Spectrum Disorder from a 3D walking-video (gait) dataset. The pipeline extracts gait features, uses nature-inspired metaheuristic optimization al…