University of New York Tirana

Komuna e Parisit,Tirana, Albania

Tel.: 00355-(0)4-273056-8 – Fax: 00355-(0)4-273059

Web Site Address: http://www.unyt.edu.al

Master of Science in Computer Science

2019-2020

Course	Business Intelligence and Data Mining
Instructor	Assoc. Prof. Dr. Marenglen Biba
Office	Faculty building 2^nd floor
Office Hours	By appointment
Phone	42273056 or ext. 112
E-mail	marenglenbiba@unyt.edu.al
Course page	http://www.marenglenbiba.net/dm/

Aims:

As the amount of available data increases at exponential rate, there is also a growing need to extract knowledge from this data. Data mining and knowledge discovery in databases have the goal of analyzing large amounts of data in order to extract relevant patterns from them. Most data mining tasks can be seen as classification tasks and here is where machine learning methods come into play. The course aims at presenting material on data mining tasks through advanced methods of machine learning. It will provide a solid foundation for IT professionals/ academics interested in the theory and practice of data mining.

Learning Outcomes:

At the end of the course the student should be able to:

1. Understand principles of data mining and machine learning.
2. Understand and describe the main models for data analysis and the differences among them.
3. Assess raw input data, and process it appropriately to provide suitable input for a range of data mining facilities/algorithms.
4. Critically evaluate and select appropriate data-mining facilities/ algorithms/ models and be able to apply them and interpret and report the output appropriately.
5. Understand and apply machine learning algorithms to data mining tasks.

Content:

Introduction to data mining and machine learning

Inductive learning

Decision trees

Rule induction

Instance-based learning

Bayesian learning

Neural networks

Support vector machines

Other machine learning models

Engineering data mining tasks

Learning and Teaching Activities:

Concepts will be introduced in lectures and tutorials.

Assessment Details:

Methods of Assessment	Please identify the LAST item of assessment that a student sits with a tick	Grading Mode	Weighting %	Minimum Pass Mark	Word Length	Outline Details
Coursework			40	50%	4000	b) Group project Case study Covering Learning Outcomes: C,D,E.
Examination	Ö		60	50%		Covering Learning Outcomes: A,B

Is the student required to pass ALL elements of assessment in order to pass the course?

YES

Indicative Texts:

ISBN Number	Author	Date	Title	Publisher
0071154671	Thomas Mitchell	1997	Machine Learning	Mcgraw-Hill International Edition
0120884070	Witten, I and Frank, E.	2005	Data Mining: Practical Machine Learning Tools and Techniques, 2^nd Edition.	Morgan Kaufmann

Software Requirements:

Weka, Oracle Data Miner 11g.

Course Material

1. Introduction to Machine Learning and Data Mining DemoVideo

2. Inductive Learning Dataset for Weka Oracle Data Miner Setup

3. Classification Task. Decision Trees Oracle Classification and Decision Trees

4. Rule Induction. First-order Logic. FOIL. Mutagenesis Dataset Mutagenesis Paper Pharmacophore Paper Inthelex Paper

5. Instance-based Learning Clustering in Oracle

6. Engineering data mining tasks Comparing Models in Oracle

7. Bayesian Learning Naïve Bayes Papers

8. Neural Networks Neural Network Drives Vehicle Face Detection with Neural Networks Handwriting Recognition Speech Recognition

9. Support Vector Machines and Association Rule Mining Market Basket Analysis in Oracle Paper on SVMs Papers on Apriori

10. Engineering the input Anomay Detection in Oracle Papers published with previous cohorts

Scientific Papers to Read

A few useful things to know about machine learning

Machine Learning in Robotics

Machine Learning for Robot Soccer

Machine Learning for Robot Soccer 2

Machine Learning for Playing Chess

Chess Player by IBM

Data Mining in Finance

Data Mining in Stock Market Analysis

Data Mining in Bioinformatics

Data Mining for real-time streams

Sample exam questions

Project specification

Last updated on Friday, October 18, 2019, by Assoc. Prof. Dr. Marenglen Biba