General information:
Code:
UBPJO-115
Name:
Pattern recognition methods and introduction to machine learning
Profile of education:
Academic (A)
Lecture language:
English
Semester:
Spring
Responsible teacher:
prof. zw. dr hab. inż. Dzwinel Witold (dzwinel@agh.edu.pl)
Academic teachers:
prof. zw. dr hab. inż. Dzwinel Witold (dzwinel@agh.edu.pl)
Module summary

Description of learning outcomes for module
MLO code Student after module completion has the knowledge/ knows how to/is able to Connections with FLO Method of learning outcomes verification (form of completion)
Social competence
M_K001 The improvement of skills concerning public speaking. Participation in a discussion
M_K002 The development of public relations through discussions with opponents. Activity during classes,
Participation in a discussion
Skills
M_U001 Development of R-language programming skills. Completion of laboratory classes,
Engineering project
M_U002 The ability of using R-langauge plug-ins in data analytics. Completion of laboratory classes,
Engineering project,
Execution of laboratory classes
Knowledge
M_W001 The broad knowledge of algorithms and methods used in pattern recognition and data analytics. Activity during classes,
Examination,
Execution of a project,
Execution of laboratory classes,
Oral answer,
Participation in a discussion
M_W002 The knowledge of fundamental theoretical base of machine learning principles. Activity during classes,
Examination,
Oral answer,
Participation in a discussion
FLO matrix in relation to forms of classes
MLO code Student after module completion has the knowledge/ knows how to/is able to Form of classes
Lecture
Audit. classes
Lab. classes
Project classes
Conv. seminar
Seminar classes
Pract. classes
Zaj. terenowe
Zaj. warsztatowe
Others
E-learning
Social competence
M_K001 The improvement of skills concerning public speaking. + - - - - - - - - - -
M_K002 The development of public relations through discussions with opponents. - - + - - - - - - - -
Skills
M_U001 Development of R-language programming skills. - - + - - - - - - - -
M_U002 The ability of using R-langauge plug-ins in data analytics. - - + - - - - - - - -
Knowledge
M_W001 The broad knowledge of algorithms and methods used in pattern recognition and data analytics. + - - - - - - - - - -
M_W002 The knowledge of fundamental theoretical base of machine learning principles. + - - - - - - - - - -
Module content
Lectures:
  1. Basics of Pattern Recognition and Machine Learning

    1. Goals and basic terms in Pattern Recognition and Machine Learning.
    2. Interpolation and regression.
    3. Multidimensional vector spaces. Hilbert and Banach spaces.
    4. Random projection
    5. Data sampling principles and methods: (Nyquist sampling, copressive sensing).
    6. Correlation and convolution for time series and images.
    7. Non-morphological and morphological filters.

  2. Time series analysis and patterns extraction using mathmatical transformations

    1. Generating of features by using mathematical transformations
    2. Appliction of Fourier transformation in feature generation and pattern recognition.
    3. Sparse coding hipothesis.
    4. Application of wavelet transformation in pattern recognition.
    5. Hough and Radon transformations in images analysis and patterns extraction.

  3. Feature generation from patterns

    1. Features generation for images: moments, Hu moments, Zernike moments
    2. Karhunen-Loeve transformation
    3. Principal Component Analysis
    4. Singular Value Decomposition
    5. Fractal compression

  4. Basics of machine learning

    1. Data representation: vectors, long vectors,matrices, graphs
    2. Proximity definitions.
    3. Classifiers and regression – examples
    4. Curse of dimensionality and basics of Vapnik and PAC theories.

  5. Feature selection and extraction

    1. Basic algorithms for feature selection: filters against wrapers.
    2. Feature extraction using PCA and LDA procedures

  6. Simple classifiers – supervised learning

    1. Linear classifiers and learning procedures: preceptron, Fisher.
    2. Non-linear classifiers: k-NN, SVM

  7. Enseble classifiers

    1. Neural networks
    2. Boosing and bagging
    3. Ada-boost
    4. Validation of classification results: cross-validation, double cross validation

  8. Clustering – unsupervised learning

    1. Simple hierarchical clustering schemes.
    2. Complex hierarchical clustering schemes: Ward, MNN, DBSCAN, CURE, Chameleon, SNN
    3. Novel clustering schemes: afinity propagation
    4. Unsupervised learning from data – deep belief networks
    5. Feature extraction using DBNs (t-SNE)
    5. Biclustering in micromatrices analysis

Laboratory classes:
  1. Using R-language for pattern recognition

    1. R tutorial
    2. Using k-NN classifier for clasiffication of selected UCI data sets

  2. Employing PCA for multidimensional data visualization

    1. Using R PCA and LDA procedures for visualization of selected UCI data sets

  3. Multidimensional scaling vs. PCA

    Using data sets from UCI compare PCA and MDS multidimensional data visualizations.

  4. Application of modern classifiers for data classification

    UCI data used in previous lab, are classified by using SVM and Ada-boost classifiers. Comparison to the results obtained for k-NN.

  5. Clustering

    Application of various clustering schemes for clustering of UCI datasets: agglomerative clustering, k-means, DBSCAN.

Student workload (ECTS credits balance)
Student activity form Student workload
Summary student workload 103 h
Module ECTS credits 4 ECTS
Preparation for classes 15 h
Completion of practical placements 25 h
Contact hours 28 h
Examination or Final test 5 h
Completion of a project 15 h
Realization of independently performed tasks 15 h
Additional information
Method of calculating the final grade:

The final grade is calculated as the weighted sum of the grades from oral (or test) exam and the lab activities. The weights are the same and equal to 0.5. The studens with excellent (5.0) grade from lab excercises are exempt from oral exam.

Prerequisites and additional requirements:

The passed courses in Mathematical Analysis, Algebra and numerical methods.

Recommended literature and teaching resources:

1. Theodoris S and Koutroumbas K, Pattern Recognition, Academic Press, San Diego, London, Boston, 1998.
2. Strang, G. and Nguyen, T., Wavelets and Filter Banks, Wellesley-Cambridge Press, Wellesley, MA, 1996.
3. Mitra, S. and Acharya T., Data Mining: Multimedia, Soft Computing and Bioinformatics, 424pp.  J. Wiley, 2003.
4. Grossman R., L., Karnath, Ch, Kegelmeyer, P., Kumar, V., Namburu, R.,R., Data Mining for Scientific and Engineering Applications, Kluwer Academic Publisher, 2001
5. R.Duda, P.Hart, D. Stork, Patterns Classification, Wiley Interscience, 2001.

Scientific publications of module course instructors related to the topic of the module:

Additional scientific publications not specified

Additional information:

None