Machine Learning

1. Applications

1.1. Recommander Systems

1.2. Natural Language Processing

1.3. Analysis of Vision and Audio Signals

2. Mathematical & Statistical Methods

2.1. K-nearest neighbour

2.2. Bayes Network

2.3. Clustering Methods

3. Supervised Learning

3.1. Training set = trained data

3.2. correct answers

3.3. Learning based on training data and apply learnings on real data

4. Recommender Systems

4.1. content - based filtering

4.2. collaborative approach

4.3. hybrid filtering approach

4.4. implicit or explicit collection of user data

4.5. Trust based recommandations

4.6. context-aware recommandations

5. Reinforcement Learning

5.1. Learning based on external feedback given

5.2. positive actions are awarded, negative actions are avoided

6. Deep Learning

6.1. Neural Networks

7. Genetic Algorithms

7.1. Travelling Salesman Problem

7.2. Simulated Annealing

8. Unsupervised Learning

8.1. Real world data without target values, the learning algorithm tries to find patterns

8.2. focus on finding hidden patterns in data

8.3. based on common characteristics/ profles the algorithm can specify into two groups of ppl in a soical network

9. Semi-Supervised Learning

9.1. Training set with missing information used - still the algorithm needs to learn from the data

9.2. draw conclusion from incomplete data