Modul

The assessment mix of each course of this module is defined for each course separately. The final mark for the module is the average of the marks for each course weighted by the credits and truncated after the first decimal.

The student

• understands the koncepts behind Semantic Web and Linked Data technologies
• develops ontologies to be employed in semantic web-based applications and chooses suitable representation languages,
• is familiar with approaches in the area of knowledge representation and modelling,
• is able to transfer the methods and technologies of semantic web technologies to new application sectors,
• evaluates the potential of semantic web for new application sectors,
• understands the challenges in the areas of Data and system integration on the web is able to develop solutions.
• know the basics of machine learning, data mining and knowledge discovery
• can design, train and evaluate systems that are capable of learning
• carry out knowledge discovery projects, taking into account algorithms, representations and applications.

None

The focus of the module is on Semantic Web Technologies as well as machine learning and data mining methods for knowledge acquisition from large databases.

The goal of the semantic web is the meaning (semantics) of data on the web for intelligent systems, e.g. in e-commerce and to make Internet portals usable. The representation of knowledge in the form of RDF and ontologies, the provision of data as Linked Data, as well as the request of data using SPARQL. In this lecture the basics of knowledge representation and processing for the corresponding technologies and application examples are presented.

The lecture "Knowledge Discovery" gives an overview of approaches of machine learning and data mining for knowledge extraction from large data sets. These are examined especially with regard to algorithms, applicability to different data representations and the use in real application scenarios.
Knowledge Discovery is an established research area with a large community that investigates methods for discovering patterns and regularities in large amounts of data, including unstructured text. A variety of methods exist to extract patterns and provide previously unknown insights. This information can be predictive or descriptive.
The lecture gives an overview of Knowledge Discovery. Specific techniques and methods, challenges and current and future research topics in this research area will be taught.
Contents of the lecture cover the entire machine learning and data mining process with topics on supervised and unsupervised learning and empirical evaluation. Covered learning methods range from classical approaches like decision trees, support vector machines and neural networks to selected approaches from current research. Learning problems considered include feature vector-based learning and text mining.

The total workload for this module is approximately 270 hours.