A collection of machine-learning projects:
-
Unsupervised Clustering (original research): developed a clustering model by implementing Spatial Autorcorrelation Analysis to detect domains of atomic dopants in graphene.
[see: http://individual.utoronto.ca/zabet/clustering.html] -
Recommender System: application of 'Content-Based Filtering' and 'Collaborative Filtering' to predict movie ratings based on users' prior ratings. The models used include Support Vector Machines (SVM), k-Nearest Neighbors (kNN), and Linear Regression.
[see: http://individual.utoronto.ca/zabet/Recommender-Systems.html] -
Spam Filter: using Naive Bayes to classify emails as ham or spam.
[see: http://individual.utoronto.ca/zabet/spam-filter.html]