Prediction using Artificial Neural Network Algorithms
This experiment is concerned with predicting the existence of disease or not from numerical data of medical historical records. We used Logistic Regression, Artificial Neural Network, and Random Forest Classifiers for the prediction. The entire data is split into 70% training and 30% testing. Before prediction, we used principal component analysis (PCA) to reduce the dimension of the training data and maximize the variance within the data. The obtained eigenvectors (principal components) are used to project the training and testing data to obtain new transformed data with much reduction of noise. Then, the 3 classifiers are used to predict the possibility of disease or not from the PCA reduced data. All experiments are repeated 5 different times.
I am interested in projects/tasks in the aspects of machine learning, image processing, and computer vision. I work with statistical models and neural network based models. I enjoy conducting research, writing report, proposal, as well as summarizing articles. I also have expertise in business analysis, business plans and proposal, copywriting, social media marketing (in particular youtube), and graphics design.