DS810: Data Driven Decision Making

The course cannot be chosen if you have passed, registered, or have followed Applied Machine Learning for Social Science, or if Applied Machine Learning for Social Science is a constituent part of your Curriculum.

Students taking the course are expected to:

Follow a course on basic introductory statistics in parallel or to have prior knowledge of basic introductory statistics. Prior knowledge to particular software packages is not a prerequisite for being enrolled. 

• Businesses can predict future sales results by combining their customers’ preference profiles with website click-stream data, social network interactions, and location data.
• Financial institutions now leverage transactional data combined with artificial intelligence models to detect fraudulent activities in real time, significantly reducing the impact of financial fraud on consumers and the banking sector.
• Manufacturing firms are implementing predictive maintenance strategies using machine learning and IoT data, minimizing downtime and extending the lifespan of machinery and equipment.
• Emergency room physicians are able to reduce time to initial treatment and, as a result, patient mortality, by fusing aggregate patient histories with the results of up to the minute lab tests.
• Energy companies are employing predictive analytics to forecast demand and manage renewable energy sources more efficiently, contributing to a more sustainable and reliable energy supply.
• With the development of electronic health records, remote treatment, and the ability to share data online, we have an array of new healthcare solutions available. The use of mobile technologies to collect and distribute information might help significantly with the prevention and treatment of disease.

• Competently use machine learning to solve the problems based on data.
• Perform clearly articulated and informed decision making based on data driven analytics.
• Account for and discuss all phases of working with data and specific methods for generating, processing/analyzing and making informed decisions on the basis of data. The student should be able to generate clear and operable management/policy implications on the basis of these three phases.
• Identify and assess data resources relevant in social sciences.
• These abilities will be documented through the work with a particular case study, which will include: a) Selecting and applying specific methods relevant for a particular case study; b) Accessing relevant data sources and analyze them; c) Clearly outlining the academic and managerial implications of working with the specific methods to an academic and a practitioner audience.

• Data analysis: Methods to process, analyze and visualize the data
• Decision making: Methods to making informed decisions on the basis of the data analysis

• James, G., Witten, D., Hastie, T., Tibshirani, R., Taylor, J. (2021, 2nd edition). Introduction. In: An Introduction to Statistical Learning. Springer Texts in Statistics. Springer, Cham. (can be downloaded free of charge from the SDU Library)
• See itslearning for syllabus lists and additional literature references.

Total number of planned lessons: 45 

Hereof: 

Common lessons in classroom/auditorium: 45

Both frontal lectures and exercises hours are focused on the application of techniques for concurrent programming to concrete problems. The difference is in the teaching method. In frontal lectures, learning will be driven by discussions directed by the teacher, whereas in exercise hours the students will have to try applying concepts by themselves first. It is expected that activities in class will be split approximately evenly between frontal lectures and exercise.

Other planned teaching activities: 

• Reading of course material
• Reflection on methods and theoretical concepts
• Gain familiarity with the programming lanquage R
• Using R to obtain and prepare data for analysis
• Using R to analyze data