Autism Spectrum Disorder (ASD) is broadly definedas a pervasive developmental disorder characterized by a triad of impairments including social communication problems, difficulties with reciprocal social interactions, and unusual patterns of repetitive behaviour. The diagnosis of ASD remains a challenging task, which requires a set of cognitive tests and perhaps hours of clinical examinations.One of the characteristic hallmarks of ASD is the difficulty of making or maintaining eye contact. In this respect, the eye-tracking technology has come into prominence to support the study and analysis of autism. This project explores machine learning as a tool for autism detection using eye tracking. We present a comparative analysis of three approaches: Convolutional Neural Network (CNN), Deep Neural Network (DNN), and Non-Neural Network models (SVM and Random Forest Classifier).