## GoalDesign, develop, and assess a Machine Learning system. - Demonstrate a practical understanding of the key theoretical concepts of modern computational/analytic methods from machine/statistical learning, data mining, and pattern recognition.
- Identify appropriate learning methods to find relationships and structure in data.
- Apply learning methods to discover patterns in data, build predictive models, and eventually to carry out a more informative decision-making.
- Develop analytic solutions to practical problems using the R statistical programming language, transforming data into knowledge.
## RequirementsBasic statistics and linear algebra. ## Detailed program## First chunk (3 CFU, by prof. Matilde Trevisani)(This chunk is part of the 12CFU version of the cource (mainly DSSC), not of the 9CFU version) - Elements of statistical learning, regression function, model accuracy, bias-variance trade-off
- Supervised learning with linear models, model checking, model selection, qualitative predictors, regularization and extensions
- Introduction and motivating example
- Data visualization and analysis with R software
- Supervised learning
- Tree-based methods: Regression and classification trees, Random forests
- Support Vector Machines
- Bias and variance and cross validation
## Third chunk (3 CFU, by prof. Matilde Trevisani)- Supervised learning for classification
- Bayes’ classifier, KNN classifier
- Linear discriminant analysis, logistic regression, quadratic discriminant analysis
- Unsupervised learning
- Principal component analysis
- Clustering: k-means, hierarchical
- Other methods of clustering
- Text mining
- Sentiment analysis
- Features for text
- Topic modeling
- Recommender systems
- Content-based filtering
- Collaborative filtering
- Evolutionary computation
## ExamEither: - Student project + written test
- Larger written test
## Lessons timetable and course calendarThe course will start on October, 3rd for the 12CFU version (DSSC) and on October, 24th for the 9CFU version. Lessons will be held in Classroom 3B, H2bis building, in Piazzale Europa campus. ## Suggested textbooks- Kenneth A. De Jong.
*Evolutionary computation: a unified approach*. MIT press, 2006 - Jerome Friedman, Trevor Hastie, and Robert Tibshirani.
*The elements of statistical learning: Data Mining, Inference, and Prediction. Springer, Berlin*: Springer Series in Statistics, 2009. - Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani.
*An Introduction to Statistical Learning, with applications in R*. Springer, Berlin: Springer Series in Statistics, 2014.
## Course materialThe course material (slides) for my portion (Medvet, 3+3 CFU) is attached at the bottom of this page. |

Teaching >

### Machine Learning and Data Analytics/Mining 2017-2018

Subpages (1):
Student project