Data analysis for semi-structured data

Semi-structured data is one of the fastest growing kinds of data in both the public and private sector, for instance in health. Email collections with sender-recipient graphs, metadata and text content is one example. This unit will explore basic forms of semi-structured data: text, time-sequence data, graphs and multiple relations in a database. Basic machine learning algorithms for these kinds of data will be analysed and applied. Some characteristic industry problems for the application of semi-structured data will also be investigated such as cohort analysis and market-basket analysis.