Data mining is a family of methods used e.g. in recommenders and in decision support systems for prediction, for customer profiling, for classification and outlier detection. For example:

• A decision maker decides which products should be offered to Internet-customers.
• The decision maker decides, when a product will be recommended to a customer, whether the customer obtains ads and how these ads look like.
• The decision maker may be a human or an intelligent service (as embedded in a recommendation engine)

For such decisions, the decision maker uses models that captures the preferences, price sensitivity and attitudes of customers, the behaviour of customers and the similarity among customers. In this bachelor course, we discuss methods for deriving models from data. In particular, we discuss

• Classification (example applications: spam recognition, distinguishing between malignant and benign tumors)
• Clustering (example applications: customer profile learning, outlier detection)
• Association rules (example applications: market basket analysis for cross selling and up selling, recommenders)

Please use the LSF application function to register for the course.

Course materials will be provided on Moodle, see hyperlink.

(You only need to enroll for the lecture.)

Pan-Ning Tan, Steinbach, Vipin Kumar. "Introduction to Data Mining", Wiley, 2004 (Auszüge, u.a. aus Kpt. 1-4, 6-8)
Selection of scientific. papers, announced during the start of the lecture

Exam admission requires a successful completion of multiple short written assessments. Details, including number and dates of the short assessments, minimum number of points to pass and minimum number of assessements to pass, will be announced at the beginning of the semester.

English Master DKE

English Master DigiEng

Export