Data Analytics for Smart Cities

Aim

In this course, you will learn how to utilize and learn from data in order to gain insights for decisions, especially in the area of smart cities. We will examine real world examples from for example traffic management, logistics, telecommunications and crowd sensing. After completing the course, the student should be able to:

• Identify the most common statistical methods used in data analytics
• Explain the differences in characteristics between different type of data analytics methods and give examples of when they should be applied
• Understand, explain and apply relevant concepts and methods in data analytics to solve practical problems
• Use selected statistical methods for prediction, classification and decision making
• Evaluate and choose among different methods for a specific problem instance
• Use existing data sets to train and evaluate selected methods for real-world applications
• Implement methods and algorithms for data analytics in a programming language

We will mainly use the statistical software Matlab to build models and work with data.

Content

The course aims to provide knowledge in data analytics, especially for applications related to smart cities. The course will cover both supervised and unsupervised learning. The focus will be on classification and prediction, but also include clustering, anomaly detection and dimensionality reduction. Example content includes statistical inference, correlation, linear regression, logistic regression, K-nearest neighbour, support vector machines, hidden Markov models, neural networks, k-means clustering and principal component analysis.