Today with the fast development of digital technologies and advance communications a gigantic amount of data sets with complex structures called ‘Big data’ is being produced everyday enormously and exponentially. The aim of the course is to give the students insights in fundamental concepts of machine learning with big data as well as recent research trends in the domain. The students will learn about problems and industrial challenges out of several case studies in the domain. Further, the students will learn to use tools to develop systems using machine-learning algorithms in big data.
The student should after course completion be able to:
- Describe and understand the basic principles of machine learning and big data
- Demonstrate the ability to identify key challenges to use big data with machine learning
- Show the ability to select suitable machine Learning algorithms to solve a given problem for big data.
- Demonstrate the ability to use tools for big data analytics and present the analysis result
Module 1. Introduction and background: introduction is intended to review Machine learning (ML) and BigData processing techniques and its related subtopics with the focus on the underlying themes.
Module 2. Case studies: presents case studies from different application domains and discuss key technical issues e.g., noise handling, feature extraction, selection, and learning algorithms in developing such systems.
Module 3. Machine learning techniques in big data analytics: this module consists of basic understanding of learning theory, association rule learning, clustering analysis and classification techniques appropriate for development work and issues in construction of systems using Big data.
Module 4. Data analytics with tools: presents open source tools e.g., KNIME and Spark with examples that guide through the basic analysis of big data.
Related industrial challenges addressed in the course:
- Structure and evaluate the vast amount of data to make sure that it is feasible to solve the customer problem.
- Acquire new, previously unknown, knowledge from routinely available huge amount of industrial data to support effective automation, decision-making etc. in industries.
- Transform knowledge acquired from the data into machines. This knowledge can be used by automated systems in various fields and provide economic values.
Upcoming instances: October 2017 (applications closed), and ’18
- Course title in Swedish: Maskininlärning med Big Data
- Course code: DVA453 (at MdH), MDH-24112 (at antagning.se)
- Course syllabus at Mälardalen University
- Autumn 2017 instance at Mälardalen University
- Admission requirements: 120 credits of which at least 80 credits in Computer Science and / or equivalent. In addition, at least 18 months of documented work experience in software development.