Developed Machine Learning Process from data preprocessing, building different learning models, and finding more powerful threshold to predict the crime rate based on demographic and economic information among severals areas. Picked up 10 types of feature affecting seriously to the high crime area based on different measures.
Technology: Python 3.6, Scikit-learn 0.18.1.
Algorithm: Decision Tree Classifier, Gaussian Naive Bayes, Linear Support Vector Machine, Linear Regression, Ridge Regression, Polynomial Features, KNN, Polynomial SVC