Exploratory Data Mining in Multimodal data
The change from a closed institution to an open living environment for patients with late stages of dementia will give the patients more freedom in their day-to-day life. The effect of this change on the patients’ mobility, activity and interaction with others will be assessed with sensor technology as well as with survey methods. This project aims at combining data from different data sources and mining information from those.
- 2018 - 2022
- Matthijs van Leeuwen
Exploring multimodal data
Developing data mining techniques to explore multimodal data and to extract interpretable information from these data will be the main contribution of this project.
The data collected in this study is multimodal as it includes different sources, such as structured interviews and observations as well as sensor data. Combining these sources and finding links between them might help us find patterns that we would otherwise miss. Pattern mining techniques that can integrate different types of knowledge will need to be developed to achieve this aim. These techniques will be developed in close cooperation with domain experts to ensure that the found patterns are useful and interpretable to the user.
Consortium: “Dementia back in the heart of the community”
This PhD project is part of the research consortium “Dementia back in the heart of the community”. The consortium includes researchers and experts in movement sciences, geriatrics and nursing as well as experts in the field of data analysis and sensor technology. It is centered around a nursing home in the south of the Netherlands and accompanies the changes in the care management and assesses the effect of the changes on the patients and staff of the nursing home.