Analysis of Massive Datasets

Data is displayed for academic year: 2023./2024.

Lecturers

Assoc. Prof.
Marin Šilić
Assoc. Prof.
Goran Delač
Assoc. Prof.
Klemo Vladimir

Lectures

Asst. Prof.
Adrian Satja Kurdija

Laboratory exercises

Stjepan Požgaj
mag. inf. et math.

Course Description

An introduction to the analysis of large datasets. Finding similar entities. Data Flow Analysis. Analysis of links in data presented by graphs. Finding frequent gatherings. Finding groups in large datasets. Recommendation systems. Social Network Graph Analysis. Web Advertising Models. Dimensionality reduction. Scalable Machine Learning.

Study Programmes

University graduate
[FER3-EN] Data Science - profile
Recommended elective courses (2. semester)

Learning Outcomes

  1. identify and understand why a problem belongs to the Big Data category
  2. apply the MapReduce programming model when encountering certain types of problems
  3. design and evaluate a system for finding similar entities in a large data set
  4. design and evaluate a system for finding frequent sets in a large data set
  5. design and evaluate a node ranking system for a very large data set represented by a graph
  6. design and evaluate a recommendation system
  7. apply appropriate algorithms to find groups in a large set of falls
  8. apply appropriate algorithms to process data flows

Forms of Teaching

Lectures

Lecturer-driven classroom presentations of theoretical concepts.

Exercises

Examples and problem solving during lectures.

Laboratory

Software implementation of selected massive dataset analysis methods. Students individually implement given assignment in a recommended programming language or tool, and submit their solutions to automatic online evaluation.

Grading Method

Continuous Assessment Exam
Type Threshold Percent of Grade Threshold Percent of Grade
Laboratory Exercises 50 % 30 % 50 % 0 %
Class participation 0 % 10 % 0 % 0 %
Mid Term Exam: Written 50 % 30 % 0 %
Final Exam: Written 50 % 30 %
Exam: Written 50 % 100 %
Exam: Oral 100 %

Week by Week Schedule

  1. Locality-sensitive hashing (LSH), minhash and simhash algorithms
  2. Locality-sensitive hashing (LSH), minhash and simhash algorithms
  3. Graph mining
  4. Web search (PageRank and HITS)
  5. Data mining with Map-Reduce, Feature selection (filter methods, subset selection, wrapper method)
  6. Data stream mining
  7. Data stream mining
  8. Midterm exam
  9. Time series and sequences mining
  10. Collaborative filtering and recommender engines
  11. Clustering algorithms for large datasets (BFR, CURE)
  12. Sampling, filtering and estimating data stream moments
  13. Large-scale algorithms for mining frequent item sets (Apriori, PCY, SON)
  14. Detecting communities in large graphs (Girvan-Newman, Affiliation-Graph Model)
  15. Final exam

Literature

(.), Jure Leskovec, Anand Rajaraman, Jeffrey David Ullman (2014.), Mining of Massive Datasets, Cambridge University Press,
(.), Michael Manoochehri (2013.), Data Just Right, Addison-Wesley,
(.), Jiawei Han, Jian Pei, Micheline Kamber (2011.), Data Mining: Concepts and Techniques, Elsevier,

For students

General

ID 222942
  Summer semester
5 ECTS
L1 English Level
L1 e-Learning
45 Lectures
0 Seminar
0 Exercises
5 Laboratory exercises
0 Project laboratory
0 Physical education excercises

Grading System

88 Excellent
75 Very Good
63 Good
50 Sufficient

Similar Courses