Erinevus lehekülje "Data Mining (ITI8730)" redaktsioonide vahel
| 5. rida: | 5. rida: | ||
Old code for this course is IDN0110 | Old code for this course is IDN0110 | ||
| + | '''Dear students, There were a problem with registration for the exam for a particular date. The problem is now fixed and besides 17.01 also 18.01 and 7.01 are available. Please accept my apology for the inconvenience. ''' | ||
Taught by: Sven Nõmm | Taught by: Sven Nõmm | ||
Redaktsioon: 3. jaanuar 2022, kell 11:15
Fall 2021/2022
ITI8730: Data Mining and network analysis
Old code for this course is IDN0110
Dear students, There were a problem with registration for the exam for a particular date. The problem is now fixed and besides 17.01 also 18.01 and 7.01 are available. Please accept my apology for the inconvenience. Taught by: Sven Nõmm
EAP: 6.0
Lectures: Tuesdays 14:00 - 15:30 SOC-414
Labs (practices): Thursdays 16:00 - 17:30 ICT-403
Link to join MS Teams https://teams.microsoft.com/l/channel/19%3a2PRNmKxRN9GR2oG68vo3_-25RYYTxAbZrA5dJ0YfoAA1%40thread.tacv2/General?groupId=9dde2da7-1d60-49a4-ac28-2d065338e369&tenantId=3efd4d88-9b88-4fc9-b6c0-c7ca50f1db57 It is advisable to use MS Teams client application and log in with TalTech account.
To join the course in talTech Moodle please use code "UseR!!!"
Consultation: by appointment only Please do not hesitate to ask for appointment!!! For communication please use the following e-mail: sven.nomm@taltech.ee
Prerequisites to join the course
Students are expected to be familiar with the foundations of Calculus, Linear algebra, Probability theory and Statistics and possess the knowledge of at least one programming language.
Overview
The course aims to provide knowledge of theory behind different methods of data mining and develop practical skills in applying those methods on practice. Is is spanned around four "super problems" of data mining:
- Clustering
- Classification
- Association pattern mining
- Outlier analysis
Main topics of the course:
- Data types and Data Preparation
- Similarity and Distances, Association Pattern Mining,
- Cluster Analysis, Classification, Outlier analysis
- Data streams, Text Data, Time Series, Discrete Sequences,
- Spatial Data, Graph Data, Web Data, Social Network Analysis
Evaluation
- 3x mandatory closed book tests. Each test gives 10% of the final grade. One make-up attempt for each test.
- 3x mandatory home assignments (Computational assignment +short write up.) Each assignment gives 10% of the final grade. Late (after deadline) assignments are accepted with penalty of 10% for each day except Saturdays and Sundays.
- final exam (gives 40 % of the final grade): Written report on assigned topic + discussion with lecturer.
Exam prerequisites: All 3 closed book tests are accepted (graded as 51 or higher), all 3 home assignments are accepted (graded as 51 or higher).
Home assignments, code examples, data files and useful links will be distributed by means of Moodle environment. Course enrollment process in Moodle TBA.
Lectures
Week 1 Distance function
Week 2 Cluster analysis
Week 3 Cluster analysis
Week 4 Outlier analysis
Week 5 Classification
Week 6 Closed Book I Test 05.10.2021
Home assignment defense 07.10.2021
Week 7 The lecture on 12.10.2021 will be given in ONLINE mode only
General analysis of the results of closed book test 1, continuation of the lecture 5, interesting discussion of a few recent trends.
Week 8 Classification
Week 9 Regression
Week 10 Associative pattern mining
Make-up test 1 Online only; November the 4th 16:00
Week 11 Tests
09.11.21 14:00 defense of the home assignments (hybrid). Online options are also available in the evening! 11.11.21 Opened book test online!
Week 12 Similarity and distance II
Week 13 Mining Time Series and Mining Data Streams
NB!!! 23.11.2021 Lecture will start 14:15
Week 14 Mining text data
Week 15 Graph data mining and social networks analysis
Week 16 Home Assignment defense on 14.12.2021 and Open Book Test 3 on 16.12.2021
No lecture or practice streaming this week!