Course plan

Save favourite

View syllabus
Syllabus:

Computer Engineering MA, Data mining, 7.5 Credits


General data

Code: DT044A
Subject/Main field: Datateknik
Cycle: Second cycle
Credits: 7.5
Progressive specialization: A1N - Second cycle, has only first-cycle course/s as entry requirements
Answerable department: Department of Information and Communication Systems
Answerable faculty: Faculty of Science, Technology and Media
Established: 3/17/2014
Date of change: 11/24/2014
Version valid from: 7/1/2014

Aim

The student should develop a basic understanding of current machine learning techniques for mining large quantities of data. The student should develop skills in finding interesting patterns and building prediction models by explorative data analysis using data analysis tools such as R, Weka or Orange, and preparing input, interpreting output and critically evaluating results. The student should show an ability to apply the skills in a small project in an real-world business or engineering application area such as big data visualization, business intelligence analysis, decision support systems, text/web/sensor/geo data mining, context aware applications, intelligent agents or cognitive radio.

Course objectives

The student should be able to:
• Discuss what real-world applications of data mining that are realistic and ethical
• Mine data using a tool such as the R script language, the Python Orange library, the Weka Java based tool or own implementations of algorithms
• Prepare input, interpret output and evaluate results
• Identify influential variables in a multivariate data set
• Discover patterns by association rule mining and evalute their reliability
• Develop and validate prediction models
• Follow a standard methological process reliable problem analysis, modelling and evaluation
• Apply data mining techniques on a small real-world problem

Content

• Application areas of data mining
• Data and knowledge representation (relations, attributes, sparse data, tables, decision trees, rules)
• Bayesian statistics
• Associative and sequential patterns
• Basic algorithms
• Data clustering
• Data categorization
• Data cleaning
• Data visualization
• Association rules
• Data prediction
• Laboratory exercise on the R, Orange or the Weka data analysis tool
• Project

Entry requirements

Previous studies 120 Credits, comprising:
- 30 credits in mathematics, including courses in Mathematical Statistics, 6 Credits and Mathematical Modeling, 6 Credits
- Java programming, 6 Credits
- Databases, modelling and implementation, 6 Credits

Selection rules and procedures

The selectionprocess is in accordance with the Higher Education Ordinance and the local order of admission.

Teaching form

The course may be offered as a campus course or as a web-based distance course. The student time commitment is estimated to about 400 hours.

Examination form

0.0 Credit, I101: Choice of project
Grades: Pass or Fail

0.5 Credit, L101: Laboratory exercise
Grades: Pass or Fail

3.5 Credits, T101: Exam
Grades: A, B, C, D, E, Fx and F

3.5 Credits, P101: Project presentation
Grades: Pass or Fail

The final grade is based on combined exam and project assessment.
Grades: A, B, C, D, E, Fx and F. A-E are passed grades, Fx and F failing grades.

Grading criteria for the subject can be found at www.miun.se/gradingcriteria.

Grading system

The grades A, B, C, D, E, Fx and F are given on the course. On this scale the grades A through E represent pass levels, whereas Fx and F represent fail levels.

Course reading


Required literature

Author: Witten, Frank, Hall
Title: Datamining - Pratical Machine Learning Tolls and Techinques
Edition: Third edition 2011 or later
Publisher: Elsivier

Reference literature

Author: Ganguly et al
Title: Knowledge discovery from sensor data
Edition: 2009 or later