Data Mining

The primary aim of this course is to acquaint students with the practical challenges that are encountered when solving real-world data mining problems. By allowing them to face the complete picture of going from business-level problem formulation, through the analysis of amount and quality of available data, as well as selection of suitable algorithms and evaluation of obtained results, we expect students to learn not only about the advantages of various machine learning methods, but also about their limitations. The main focus in this course will be on applying, in practice, knowledge and concepts from a number of previous Artificial Intelligence courses and on understanding the applicability of data mining and machine learning. In addition to presenting a number of unsupervised learning algorithms, this course will provide principles and practices for all the steps surrounding the "narrow" application of AI, such as obtaining, synchronising and maintaining the data, cleaning and structuralising the data, posing questions that are both useful and answerable, analysing and evaluating results. It is important that students are prepared to handle data that comes in many different forms (free text, databases, sensor readings, transaction history, images and video, etc).