
|
Department of
Computational
Perception
|

|
Learning from User-generated Data
344.046, KV, 3h (4.5
ECTS), SS 2014
Univ.-Ass.
Dr. Peter Knees
Assist.-Prof.
Dr.
Markus Schedl
Time:
Tuesday,
13:45 – 15:15
Start: Tuesday, March 11, 2014
Location: S3 048
This class will be taught in English.
Class Objectives and Content
Students will acquire basic
knowledge for dealing with used-generated data in the fields of machine
learning and pattern recognition. They will learn about sources and
methods for
data extraction from the web and social media as well as techniques for
processing
this data. The students' skills will be further developed through a
practical
project involving work with real-world data.
The main topics covered include:
- sources for user-generated data
- mining and analysis of web content and structure
- mining and analysis of social media
- mining of user behaviour and feed-back (explicit and implicit)
- recommender systems
- strategies for personalisation of content and user interfaces
- context-aware search, retrieval, and recommendation.
Practical Exercise
Using
real-world, user-generated data, students will conduct a pattern
classification task that requires data extraction, data analysis,
feature definition, training of various classifiers, and systematic
experimentation.
last edited by pk on Feb 9, 2014