The Data Science Research Group focuses on core data science research, as well as on applications where data science can provide insights for decision making. We formulate novel data science problems and develop algorithmic methods and methodological workflows.
Illustration: Sergey Nivens/Mostphotos.
The Data Science Research Group is particularly focusing on exploiting huge amounts of data to enhance data-driven decision-making in areas such as healthcare and integrated vehicle health management. We put special emphasis on sequential and temporal data, as well as on text.
In addition, we are interested in building methods and workflows for explainable machine learning. We aim to describe the opaque machine learning models to humans, and to provide explanations and motivations for each decision of the models. Our main goal is to provide scalable and distributed solutions for maintaining good trade-offs between predictive performance and explainability.
Our methods and solutions are motivated by real-world applications and use cases. The group has particular expertise in mining and model understanding from healthcare and medical data sources. We have also established a strong expertise in predictive maintenance and integrated vehicle management. Finally, we are interested in financial data, environmental data, and data emerging from immersive technologies, such as virtual reality (VR).
The A3S project aims to create an AI-powered asthma app with a spirometer, using explainable AI to analyze patient data and guide decisions. It focuses on transparent, person-centered care to help people manage asthma and reduce health risks.
This project aims to make AI fair, transparent, and just, ensuring it benefits everyone in society. By addressing biases in healthcare and education, we’re working to build AI systems that serve society equitably and responsibly.
Digital twins and artificial intelligence are two of the key driving technologies of the fourth industrial revolution. This project connects the two fields in an industry–academia collaboration between Stockholm University and Atrium Ljungberg.
The shift towards renewable energy comes with technical challenges. We investigate how AI-algorithms can be used in the electric networks of our cities for smart coordination and control of the more volatile, renewable energy resources.
AI and data-driven autonomous decision-making will play a leading role in the digitalization of critical infrastructures for electricity, water, and transportation. This project explores our future smart societies.
The aim of this EU project is to increase food safety for citizens. Today there is a wide array of data sources holding crucial information about the food that we eat. The problem is that these sources are heterogeneous – and sometimes hidden. We explore how data can be mined, aggregated and analysed using AI.
The goal of thid cross-disciplinary collaboration is to design and implement a novel data management and analytics framework for medical data sources. The focus is on explainable machine learning methods as well as on legal and ethical aspects of the predictive models.
Federated Reinforcement Learning is an emerging research field which allows multiple AI systems to learn together – without sharing their private data. In this project, we explore how scalable systems can be built.
How can cities become smarter? Our Baltic Sea university consortium is developing scalable, secure and smart solutions by integrating IoT, edge, and cloud computing. The goal is to improve sustainability, energy use, and service delivery in diverse urban environments.
The OrganoFeed project intends to make lab-grown mini-organs – organoids – more reliable for research. Using AI and microfluidics, it fine-tunes the organoid environment in real time, reducing errors. This could improve drug testing and even help replace animal experiments.
With better descriptions of a patient’s state and history, more efficient recommendations can be provided. We explore how AI tools can be put to practical use in healthcare. We focus on complex and multimodal data and use cases such as COVID-19 public health interventions or patient phenotyping for adverse drug events, sepsis, or cancer.
Less accidents on the road, and more operational uptime. That is the expected outcome of this research project which uses data from trucks to develop new machine learning models. The models will let us know when maintenance is needed – before the vehicle breaks down.
In this project, we investigate the next-generation distributed AI and Machine Learning algorithms in complex networks. Our aim is to find more sustainable solutions.
In this project, we leverage advanced AI to develop smart power converters. They stabilize the electricity grid, enabling a secure, resilient and renewable-rich power supply for a climate-neutral society.
In order to advance intelligent data analysis, there is a need for novel and potentially game-changing ideas. This assumption is the foundation for the IDA symposium which in 2024 was organised at Stockholm University. Explainable AI was one of the important themes.
A new wave of wearable devices will collect a mountain of information on us. And there are privacy implications, writes Luis Quintero in The Conversation.
Sweating, facial expressions and increased heart rate. Our bodies send signals about our emotions – signals that can be picked up by sensors. The input can then be used to design our next workout, meal or learning experience. Luis Quintero’s research provides a sneak peek into the future.
Data is available in abundance. However, to utilize this raw material we need to sort it so that patterns and trends become visible. Zed Lee develops algorithms that are useful for industry as well as healthcare.
AI technology is breaking new ground in all areas, not least in medicine. What can we expect in a near and distant future? PhD students and supervisors from five countries gathered at Stockholm University to discuss their projects and learn from each other.
