Data Mining and Knowledge Discovery

professor: prof. Nada Lavrač

professor: prof. Bojan Cestnik

assistant: dr. Martin Žnidaršič

assistant: dr. Petra Kralj Novak

Courses

• Information and Communication Technologies (ICT2): Data and Text Mining / Podatkovno in tekstovno rudarjenje (20 ECTS)
• Information and Communication Technologies (ICT3): Data Mining and Knowledge Discovery / Podatkovno rudarjenje in odkrivanje zakonitosti (10 ECTS)
• Statistics PhD program: Podatkovno rudarjenje in odkrivanje zakonitosti v podatkih (5 kt)

Scheduled lectures and course materials in school year 2017/18

Practice materials: dr. Petra Kralj Novak

• Practice notes:
  • 7.11.2017 slides (.pdf) , 6/page (.pdf)
  • entropy and information gain (.pdf)
  • decision trees (.pdf)
  • 8.11.2017 slides: corrected AUC example (.pdf) , 6/page (.pdf)
  • naive Bayes (.pdf)
  • 14.11.2017 slides: corrected AUC example (.pdf) , 6/page (.pdf)
  • association rules (.pdf)
• Hands on Weka
  • Hands on Weka Part 1 (.pdf), 6/page (.pdf)
  • Hands on Weka Part 2 (.pdf), 6/page (.pdf)
  • Hands on Weka Part 3 (.pdf), 6/page (.pdf)
  • datasets (.zip)

Book clubs:

1. Basic book club: Max Bramer: Principles of Data Mining (2007)
   • We will study the book: Max Bramer: Principles of Data Mining (2007) and discuss it in three sessions
   • Chapters 1-5
   • Chapters 6-10
   • Chapters 11-15
2. Advanced Book club: Charu C. Aggarwal : Data Mining: The Textbook
   • We will study secected chapters from the book: Charu C. Aggarwal : Data Mining: The Textbook and discuss it in four sessions
   • Chapters: 1,2,3,4,6,8,10,(maybe 11)

Course requirements ICT2, data mining part:

• Attending lectures
• Attending practical exercises : Theory exercises and hands-on (intro to WEKA by Petra Kralj Novak)
• Collaborating in one of the book clubs
• Written exam
• Seminar: written part, oral presentation, both in English

1. Written exam
   • 60 minutes of time
   • 4 tasks: 2 computational (60%), 2 theoretical (40%)
   • Literature is not allowed
   • Each student can bring one hand-written A4 sheet of paper and a hand calculator
2. Data mining seminar
   • Data analysis of your own data (Weka recommended)
   • One page seminar proposal on written exam day
   • Deliver a 4 pages written report (printed and electronic copy) in Information Society paper format
     on seminar presentations day (use paper template and guidelines)
   • Oral presentation of seminar results (10 minutes for presentation + 5 minutes discussion, use slides template)

Course requirements ICT3

• Attending lectures
• Attending tutorial: Intro to ClowdFlows by dr. Martin Žnidaršič
• Collaborating in one of the book clubs
• Written exam
• Seminar: written part, oral presentation, both in English

1. Written exam
   • 60 minutes of time
   • 4 tasks: 2 computational (60%), 2 theoretical (40%)
   • Literature is not allowed
   • Each student can bring one hand-written A4 sheet of paper and a hand calculator
2. Data mining seminar from advanced data mining topics
   • Data analysis of your own data in Clowdflows or by using other data mining toolboxes
   • One page seminar proposal on written exam day
   • Deliver a 4 pages written report (printed and electronic copy) in Information Society paper format
     on seminar presentations day (use paper template and guidelines)
   • Oral presentation of seminar results (10 minutes for presentation + 5 minutes discussion, use slides template)

Examples of data mining seminars:

• Janez Bucik (.pdf)
  Microsoft stock quotes dependency analysis
• Valentin Koblar (in Slovene .pdf)
  Napoved menjalnega tečaja ameriškega dolarja na podlagi menjalnih tečajev tujih valut

Ideas for seminars

1. Analyze some data where you are the domain expert, use at least two algorithms
2. Find some interesting data to analyze, possible sources:
   • Statistični urad Republike Slovenije
   • Ljubljanska borza d.d.
   • Banka Slovenije
   • World Health Organization
   • Center for Climatic Research, University of Delaware
   • Vrhovno sodišče Republike Slovenije - sodna praksa

Templates

• presentation template (.pot)
• paper template (.doc)
• paper guidelines (.doc)

Useful links

• Weka
• ClowdFlows

Link to last year's web page - Data Mining and Knowledge Discovery 2016/2017
Last update: 20180111