Fatıh Unıversıty,

Computer engıneerıng

CENG 574 Data mınıng Fall 2011

 

ANNOUNCEMENTS

 

 

 

Instructor: Atakan Kurt, Room EA306, ext 5513, akurt@fatih.edu.tr

Office Hour: by appointment.

Assistant: Murat Gezer, E Blok E 107, ext: 5522, mgezer@fatih.edu.tr

 

Web page: http://www.fatih.edu.tr/~akurt/courses/ceng574/fall2011/

Announcements, homeworks, slides etc. will be made available in the web page.

 

Description:

The objective of the course is to learn the fundamental techniques and methods  in Data Mining. We will cover important techniques in supervised learning, unsupervised clustering and association rules. We will also cover the recent applications of data mining techniques to web, spatial and temporal domains. Weka will be be used for practical applications of data mining techniques in the class

 

Grading:

6 quizess from 6 chapters. Quiz 5 (classification) and quiz 6 (clustering) will count for midterm and the final exams respectively.

 

Textbook:

Data Mining, Intoductory and Advanced Topics

Margareth Dunham, Prenctice-Hall, 0-13-088892-3, 2003

Supplemental material:

Data Mining, Concepts & Techniques

Jiawei Han, Micheline Kamber, Morgan Kaufmann, 1-55860-489-8, 2001

 

Contents

·        Introduction to Data Mining Concepts

·        Related Concepts in Data Mining

·        Data Mining Techniques

·        Classification: Regression, Bayesian Classification, Distance based Algorithms, K-nearest neighbors, Desicion tress, Neural Networks, rule-based Algorithms

·        Clustering: Similarity-distance metrics, Agglomeratve algorithms, Divisive algorithms, Partional Algoirthms, Minimum spanning Tree, Squared Error Clustering, K-means, Nearest Neighbor, PAM, Bond Energy, clustering with NN and GA

·        Association Rules: Apriori Algorithms, Sampling algorithm, Partioning, generalized, multilevel, quantitative association rule algorithms, mesuring quality,

·        Web mining:Web content mining, structure mining and usage mining.

·        Spatial Mining: Spatial data, and spatial DM primitive, Spatal rules, spatial clustering and spatial clustering algorithms.

·        Temporal Mining: Modeling events, Time Series, pattern detection, Sequences, Temporal Association Rules.