Implement different type of machine learning models to deal with practical problems, majorly using Python
- Building Price Models
- Classifiers like Bayesian classifiers, decision trees, SVM can be used for prediction, but may not be the best choices for predictions about numerical data with many different attributes.
- Check ReadMe_PriceModels.txt for all the details: https://github.com/hanhanwu/Hanhan-Machine-Learning-Model-Implementation/blob/master/ReadMe_PriceModels.txt
- Advanced Classification for MatchMaker dataset
- Dating website looks really magical, they match people based on the info each individual provided, the generated dataset is called MatchMaker dataset, which contains both numerical and nominal data.
- I have practiced advanced classification models like SVM, linear classifiers and kernel methods in this part of code.
- Finally, I have parsed friends data through Facebook Graph API, then did data preprocessing and finally used SVM to do friends prediction.
- Check ReadMe_AdvancedClassification.txt for all the details: https://github.com/hanhanwu/Hanhan-Machine-Learning-Model-Implementation/blob/master/ReadMe_AdvancedClassification.txt
- Searching and Ranking
- I am creating a basic search engine in this part of code, and there are 4 steps: crawling, indexing, ranking, building neural network for ranking queries
- Check ReadMe_SearchRanking.txt for all the details: https://github.com/hanhanwu/Hanhan-Machine-Learning-Model-Implementation/blob/master/ReadMe_SearchRanking.txt
- Optimization
- Check ReadMe_Optimization.md for all the details: https://github.com/hanhanwu/Hanhan-Machine-Learning-Model-Implementation/blob/master/ReadMe_Optimization.md
- Decision Tree
- Check ReadMe_DecisionTree for all the details: https://github.com/hanhanwu/Hanhan-Machine-Learning-Model-Implementation/blob/master/ReadMe_DecisionTree.md