The MSc BI programme is still new and the structure of courses and their content have been changed and will continue to change slightly, in order to constantly improve the scope and adapt to the demands in the field of Business Intelligence outside the university.
MSc. Business Intelligence in Aarhus BSS: 2023/2024 Structure
Prerequisite courses (first semester)
Business Intelligence (BI) uses data to create business value for decision-makers. Successful implementation of BI, therefore, requires an understanding of data management as well as the skills for presenting the data to the end-user in an easily accessible format.
Database management is concerned with the structuring of data in order to efficiently meet the organisation’s data needs. This structuring is important as the various operational systems in a typical company create vast amounts of data, and external data sources are expanding rapidly. The student will therefore learn how to use data management systems and get hands-on experience with writing SQL.
When the data is adequately structured, it is vital to transform this data into information and to deliver this information in the right format to the right user. To achieve this end, the student will learn how to report and visualise information using the R Shiny software package.
Both data management and reporting are important elements in the Enterprise Information Infrastructure. This course covers the foundational tools and skills the student will need to be a skilled business intelligence analyst.
Data are at the heart of many core business processes in a modern company. Data are generated by transactions in operational systems and are being warehoused to become part of the corporate memory. In industries such as banking, insurance, media, telecommunications, retailing, and e-commerce, the amount of data is often so vast that companies complain about “drowning in information but starving for knowledge”.
Machine learning is about learning from data. A machine learning system is trained rather than explicitly programmed. The user presents the system with several examples relevant to a task, allowing the system to identify statistical structure in these examples, which then enables the system to come up with rules for automating the task. In other words, we use data to build a prediction model, or learner, which enables us to predict the outcome for new unseen objects.
Data-driven business and evidence-based decisions are significant competences for companies and managers. The ability to lead through data, information and knowledge is significant to any company. Business Intelligence is an umbrella definition that is concerned with the use of data to create valuable insights for decision-making and to create business value. Building a data-driven enterprise can be a daunting task, but it is imperative if a company wants to compete (and survive) today.
This course is designed to give a solid theoretical and applied background to graduate students in forecasting. Students are expected to have taken the bachelor’s course Quantitative Research Methods, or an equivalent course that covers regression analysis with a good understanding of the statistical methodology used. The course will not only be a methodology course but equally an applied course in that students will develop skills to approach business life situations critically, evaluate and communicate their findings with ease. The applications that immediately follow the theoretical topics to be taught will cover different business topics including sales planning, product demand analysis, marketing and advertising.
Specialization courses (second semester)
The purpose of this course is to provide students with a strong conceptual and technical foundation in the area of customer analytics defined as the use of customer databases to enhance marketing productivity through more effective acquisition, retention and development of relationships with customers. The customers can be either current customers or prospective customers. The course enables the students to understand, evaluate and use analytical methods such as Markov chain, structural equations models, and directed acyclic graphs (Bayesian networks) to analyse the content of customer transaction databases as well as to construct surveys and experiments for collecting new data with the purpose of supplementing a firm’s existing knowledge about customers.
The purpose of this course is to expand the foundation provided in Machine Learning for Business Intelligence 1 to advanced non-linear machine learning techniques for modelling and prediction from data. The course enables students to understand the models, intuitions, and strengths and weaknesses of the various approaches without diving into the technical details behind the methods. In particular, the overall goal is to empower the students to become informed users of advanced machine learning techniques, thereby preparing them for their future roles as data scientists or business intelligence consultants.
The course consists of two parts. Part 1 covers tree-based methods (bagging, random forests, and boosting) and support vector machines. Central here is the practical application of these methods to a wide variety of (mostly) high-dimensional datasets. Students will learn how to tune hyperparameters, build ensemble models, and employ useful tools for model validation. Students are provided with broad knowledge and practical experience with several central R libraries for advanced machine learning, e.g., RandomForest and e1071.
Part 2 covers neural networks, including deep learning. Students will learn best practices for training neural networks for computer vision and natural language processing. Part 2 teaches students how to build and choose between state-of-the-art architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and Transformers. Students are taught how to handle real-world image and text data, and explore strategies to prevent overfitting, including augmentation and dropout. Students are provided with a broad knowledge of and practical experience with TensorFlow through Python.
Business is facing a rapidly growing information stream including new data types. Meanwhile, advancements in the development of analytical algorithms and computational power expand the boundary of possible analytics. On top of this, there is a growing need for people who are able to combine high technical and mathematical skills with business understanding and abilities to solve complex problems. The overall purpose of this course is therefore to enable students to solve complex real-world business problems using data.
This course will teach students methodologies and processes for creating business value through data. The full end-to-end perspective will be in focus, from discovering and defining relevant business problems to generating actual value for companies.
The course will let the students combine knowledge and tools for data science projects to solve business problems from end-to-end. In parallel, the course will focus on the theoretical understanding of data at different levels of refinement.
In the third semester you can choose elective courses within your areas of interest. The courses can either be taken at School of Business and Social Sciences during the semester, at the Summer University or at one of our more than 300 partner universities abroad. You can also participate in internship programmes either in Denmark or abroad.
The fourth semester is devoted to the final thesis. You may choose the topic of the thesis freely and so get a chance to concentrate on and specialise in a specific field of interest. The thesis may be written in collaboration with another student or it may be the result of your individual effort. When the thesis has been submitted, it is defended before the academic advisor as well as an external examiner.