This repository is for Machine learning projects or Kaggle competitions.
- Digit Recoganizer (Kaggle)
- Titantic (Kaggel)
- SVD (image compression experiment)
- Neural Network/Back Propagation Java implementation
- Random Forest implementation and human written digits recoganization
- (t-SNE) is a technique for dimensionality reduction that is particularly well suited for the visualization of high-dimensional datasets.
- Implementation of EM (expectation maximization) algorithm for A two-coin-flipping experiment (http://ai.stanford.edu/~chuongdo/papers/em_tutorial.pdf)
- PLSA java implementation to find topics and concepts from weibo. It is based on EM algorithm. A reference is http://blog.jqian.net/post/plsa.html
- Baum-Welch java implementation to train HMM (hidden Markov model) with only observed sequences, a.k.a Baum-Welch is un-supervised learning algorithm.
This tries to find topics/concepts from a list of weibos.
t-SNE 