Pattern Recognition

Recommended Prerequisites

It is recomended to have knowledge in Algebra and English   

Teaching Methods

Interactive lectures, with many applications and examples.    

Learning Outcomes

Acquire knowledge in methods, algorithms and tools in multivariate analysis.  

Work Placement(s)

No

Syllabus

  Basic Statistics

Random variables and probability distributions. Joint distributions. Moments, variance and standard deviation. Covariance. Statistical tests. Bayes formula and Bayesian inference.

• Methods of correlation and time series

Covariance, correlation and regression. The variance-covariance matrix. Series: autocovariance and autocorrelation.

• Cluster analysis

Definition. Distances, similarity. Hierarchical techniques. Sharing-optimization techniques. Other methods.

 • Analysis of principal components and factors

Interpretation. principal components in two dimensions and n. Applications. Analysis of factors: determination and rotation.

 • Supervised Pattern Recognition

Decision rules. Linear discriminant analysis. The k-nearest neighbor method.

 • Applications

Chemical analysis, fingerprinting, physical traces, facial recognition, forensic anthropology  

Head Lecturer(s)

Alberto António Caria Canelas Pais

Assessment Methods

Final assessment
Exam: 100.0%

Bibliography

R. Lyman Ott, An Introduction to Statistical Methods and Data Analysis, Duxbury (Pacific Grove , 2001)

R. Wehrens, Chemometrics with R, Springer (Heidelberg, 2011)

R.G. Brereton, Chemometrics : Data Analysis for the Laboratory and Chemical Plant, Wiley (Chichester, 2003).

D.L. Massart, B.G.M. Vandeginste, S.N. Deming, Y. Michotte e L. Kaufman, Chemometrics: a textbook, Elsevier (Amsterdam, 1988).