Text Mining

COURSE OBJECTIVE
You will get acquainted with the possibilities and problems of automatic analysis of natural language by computers. Students will obtain practical knowledge; they will learn to use existing technology and experience the obstacles and options of the domain. They will learn about the theories behind language technology and its connection to artificial intelligence, linguistics and semantic web. The students will choose a project themselves in which they apply the learned technologies, evaluate its results and communicate their findings through a report.

COURSE CONTENT
It is estimated that about 80% of knowledge is captured in language: think of news, wikis, social media and handbooks. Searching for information is also largely done through language. The amount of information is too large for humans to oversee, which is why technologies are developed to access and use this information more efficiently.

Text Mining is a promising research domain whose goal it is to extract structured information from unstructured natural language. This is a big challenge as human language is a rich and complex medium that is to be understood in the context of social human interaction. Therefore, language technology analyses language on different levels: the grammatical level (e.g. word types and syntax), and the semantic level (e.g. entities, events, opinions). During the course you will learn how this information is coded in text and how you can extract and present it using computers.

FORM OF TUITION
Lectures and labs.

TYPE OF ASSESSMENT
Assignments and exam:
50% final assignment (group);
50% exam.

None of the grades can be lower than 5.5 to pass the course. Attendance at the final assignment presentation session is mandatory and all but one of the practical assignments need to be passed.

RECOMMENDED BACKGROUND KNOWLEDGE
Information Retrieval and Python