AI-based Decision Support I
The lecture deals with the fundamentals of machine learning, especially supervised learning. The goal is to create prediction models and therefore design a full analysis process from business goal to deployment. We will deal with basic steps of any data analysis process using the CRISP-DM process model: business understanding, data understanding, data preprocessing, modeling, evaluation and deployment. The students will learn how to deal with each phase and to question and analyze results before deploying them in a productive environment. Students will therefore learn how to design AI-based decision support systems from front to back that can tackle a practical forecasting problem. As a result, students will acquire analytical thinking qualifications and critical thinking when evaluating the models. Besides preparing students for the job of a data scientist, the course will prepare you for advanced analytics courses (AI-based Decision Support II in the summer term), seminars and scientific projects that you may take during your course of study. In addition, the on-demand online exercises will enable the students to gain competencies in python data science programming.
The syllabus of the course is planned as follows:
(1) Introduction, (2) Data Exploration, (3)Preprocessing, (4) Modelling (Regression, Neural Networks, SVM ), (5) Evaluation of Machine Learning Algorithms.
The lecture is supported by DataCamp for Classroom exercises.
Literature: Bishop (2006) - Pattern Recognition and Machine Learning. Springer.
E-Learning Link for WiSe 2022/23: https://elearning.ovgu.de/course/view.php?id=13271