Identification of Autism Spectrum Disorder using ML

The goal of this report is to apply machine learning algorithms to classify autism spectrum disorder (ASD) patients and typically developing (TD) participants using fMRI data from ABIDE dataset. SVM and KNN were used for classification purpose. Multi-layer perceptron classifier was also used for comparison. I used a cross-validation grid search to fine-tune the hyperparameters for each classifier. Finally, a stacked ensembled model was used with the tuned hyperparameters of the classifiers. More information can be found in report.pdf file.