Data Science Research
The Data Science arena includes several research groups working on a large variety of basic research methods and applications. On this page we will post a monthly highlight of our research.
Data Science and Health
An explosion of AI related research is currently revolutionizing the medical sciences, ranging from the prediction of epidemics to the monitoring of patients. One of the use cases where machine learning can improve our quality of life is with the analysis of ECG readings. The classifiers developed from a collaboration between researchers in machine learning at the Department of Information Technology and medical researchers from Brazil can be useful in countries where it may be difficult to have a cardiologist promptly available in case of emergency.
From PDEs to neural networks (and back)
Deep neural networks have become one of the success stories in modern machine learning, providing state-of-the-art solutions in areas such as computer vision and natural language processing. However, we still have limited knowledge of these models. A mathematical theory of deep learning can help us understand how they function, potentially improving the way in which we use them. At the same time, deep neural networks can be used to address hard mathematical problems.
Data Science in Telecommunications
Telecom companies are working to deploy the fifth generation (5G) cellular networks in Sweden to augment the existing networks. Much has been said about the increased speeds and reduced latency of these new networks, the first of which was launched in Stockholm in 2020. But how can we make optimal use of the changes coming with this new generation of networking technologies?
Data Science and Contagious Diseases
As early as the 1850s Data Science has played an important part in the study of diseases. The IT department at Uppsala University is continuing the line of research set out by pioneers like John Snow using state of the art modeling tools. One of their current projects focuses on the spread of infectious diseases in the Swedish cattle population.
Social Data Science
Many of our social interactions are observable online, for example from the digital footprints we leave behind us when we interact on social media sites. Within appropriate legal and ethical limits, this data can be used to learn more about social structure and processes, from power relations to information dissemination. However, this is also extremely complex data to model and analyse, including networks, text, images and temporal annotations and thus requiring beyond-state-of-the-art methods.
Data Science and Cloud Computing
The complexity of data produced by scientific experiments often requires the application of knowledge and methods from different areas. For example, ideas from scientific computing, machine learning and distributed systems research can be integrated to find new ways to do computation-driven study of natural phenomena.
Cancer Detection with Data Analysis
Data Science can be used to help cancer specialists to perform difficult, time-consuming, expensive and error-prone analyses. But besides having a high performance, automated cancer detection methods should also be considered trustworthy by both doctors and patients to be useful in practice.