Machine Learning Overview

Online Resources

• Journal: http://jmlr.csail.mit.edu/papers/
• Blog: http://hunch.net/
• Surveys: http://www.mlsurveys.com/
• Tutorials: http://www.autonlab.org/tutorials/
• Lectures: http://www.cs.cmu.edu/~tom/10701_sp11/lectures.shtml
• Course: https://www.coursera.org/course/ml, notes: http://www.holehouse.org/mlclass/
• Statistics: http://onlinestatbook.com/
• Convex optimization: http://www.stanford.edu/~boyd/cvxbook/
• Information retrieval: http://trec.nist.gov/ 
• Machine learning summer school: http://www.mlss.cc/
• Kaggle competitions: http://www.kaggle.com/competitions
• MIT OpenCourseWare
• Best Papers vs. Top Cited Papers in Computer Science (since 1996)
  
• 水木社区 → 学术科学 → AI（人工智能）版
• Datasets:
  • UCI Machine Learning Repository
  • KONECT Datasets
  • Data 数据堂
  • SNAP: Network Datasets
  • Social Computing Datasets
  • Trustlet Datasets
  • Yahoo!Music Dataset
  • Twitter Links
  • Kevin Chai's Dataset
  • BooK Crossing Dataset
  • A Rich Epinions Dataset
  • Clustering Data Sets

Learning List

• Random Forest
• Bayesian Methods

Supervised Learning

• 
  

Unsupervised Learning

• Unsupervised learning refers to the problem of trying to find hidden structure in unlabeled data. Many methods employed in unsupervised learning are based on data mining methods that are used to preprocess data. The below organization refers to (Zoubin Ghahramani, 2004)
• Latent variable models
  • Factor analysis
  • Principle components analysis
  • Nonnegative matrix factorization
  • Independent component analysis
  • Gaussian mixture models
  • Sparse coding and dictionary learning
• Clustering methods:
  • K-means
• The EM algorithm
• Time series models
  • State-space models
  • Hidden Markov models
  • Modeling other structured data
• Graphical models
  • Undirected graphs
  • Factor graphs
  • Directed graphs
  • Expressive power
• Exact inference in graphs
  • Elimination
  • Belief propagation
  • Factor graph propagation
  • Junction tree algorithm
  • Cutest conditioning
• Learning in graphical models
  • Learning graph parameters
    • The complete data case
    • The incomplete data case
  • Learning graph structure
    • Scoring metrics
    • Search algorithms
• Bayesian models
• Approximating posteriors and marginal likelihoods
  • Laplace approximation
  • The Bayesian information criterion
  • Markov chain Monte Carlo
  • Variational approximations
  • Expectation propagation
• Other unsupervised learning methods
  • In ANN, the self-organizing map (SOM) and adaptive resonance theory (ART) are unsupervised methods. 
  • Anomaly detection methods. 

References

1. Zoubin Ghahramani, 2004, Unsupervised Learning. (more refers to: http://mlg.eng.cam.ac.uk/zoubin/reviews.html)