|Teoria||4||I semestre||Manuele Bicego|
|Laboratorio||2||I semestre||Manuele Bicego|
The course is aimed at providing the theoretical and applicative basis of Pattern Recognition, a class of automatic methodologies used to recognize and recover information from biological data. In particular, during the course the main techniques of this area will be presented and discussed, in particular linked to representation, classification, clustering and validation. The focus is more on the description of the employed methodologies rather than on the details of applicative programs (already seen in other courses)
At the end of the course, the students will be able to analyse a biological problem from a Pattern Recognition perspective; the will also have the skills needed to invent, develop and implement the different components of a Pattern Recognition System.
The course generally requires standard skills obtained from other courses of the first two years, with particular emphasis on basic notions of probability, statistics, and mathematical analysis.
The course is divided in two parts:
Theory. This part is devoted to the description and the analysis of the different methodologies for representation, classification and clustering of biological data. Moreover, there will be a more application-oriented part, which is devoted to the critical analysis of some relevant bioinformatics problems which are typically solved with classification or clustering approaches (e.g. gene expression data analysis, medical image segmentation, protein remote homology detection)
Laboratory. This part is devoted to the implementation, using the MATLAB language, of some of the algorithms analysed in the first two parts.
- Introduction to Pattern Recognition
- Data Representation
- Elements of the Bayes decision theory
- Generative and discriminative classifiers
- Elements of Neural Networks and Hidden Markov Models
- Clustering methods
- Introduction to matlab
- Data representation and standardization
- Principal Component Analysis
- Gaussians and Gaussian classifiers
- Hidden Markov Models
R. Duda, P. Hart, D. Stork Pattern Classification. Wiley, 2001
P. Baldi, S. Brunak, Bioinformatics, The Machine Learning Approach. MIT Press, 2001
A.K. Jain and R.C. Dubes, Algorithms for Clustering Data, Prentice-Hall, 1988
See the general notes on the course.