With the growing demand for data science skills in optimizing business processes, our interdisciplinary approach addresses real-world problems in various fields. As a result, MU Vienna has founded the School of Applied Data Science and launched our new BSc in Applied Data Science, capitalizing on our interdisciplinary tradition and comprehensive quantitive data-driven approach.

The School of Applied Data Science and its faculty engage in a multitude of research activities.

Web and Network Science

Huge amounts of data available on the Web and in social media allow us to analyze important social and economic phenomena such as diffusion of information, emergence of communities, or self-organization in collaborative efforts. To analyze such large datasets methods from graph theory and network science are of primary importance as the data often comes in form of relations between different entities. For example, data in social media is typically represented with social networks and the data on the Web with infromation networks. In our research we quantitatively analyze large emprirical datasets but also develop new methods for network analysis.

Analysis of Behavioral Data

We work with obsevational data and design new methods for extracting and infereing causality from observational data. As the observational studies lack a clear experimental design, we typically adopt quasi-experimental designs such as natural experiments and matching. With these methods we analyze interesting questions such as resilience of self-organized online communities, user behavior in the online discussion of controversial topics, or user eating habits.

Recommender Systems

When users are confronted with multitude of options such as millions of products to chose from, millions of goods to consume, or millions of news to read, recommender systems support users in selecting the appropriate options. By analyzing and training on the historical data of past user interactions with items, recommender systems learn useful and compact user and item representations that can be used to predict future user-iterm interactions. In our research, we work on developing new algorithms for recommender systems that can be used in so-called conversational recommendations applying tools such as large language models.

Predictive Analysis
Predicting the future is valuable for individuals and businesses, enabling strategic changes to maximize benefits or minimize potential damage. With advancements in big data, AI technologies, and predictive analytics, our School of Data Science researches how various data inputs can improve accuracy in open domain settings. While perfect knowledge of the future is elusive, our findings demonstrate how knowledge extracted from big data can guide organizations in making informed decisions. Join us to explore the power of predictive analytics in shaping a more probable future.

Visual Classification of Images

Computer vision enables computers to "see" and understand visual content, thanks to advancements in AI and deep learning. At the School of Data Science, our research focuses on applying deep learning-based visual classification to measure destination images. By training our network on tourist-related images, we can accurately classify and compare how destinations are presented visually on platforms like Instagram. This valuable insight allows destination marketing organizations to tailor their online marketing efforts to align with tourists' interests. Join us in exploring the innovative intersection of computer vision and destination marketing.