Research Support

The Story Navigator is a computational tool that supports researcher when analyzing texts. The tool gives the researcher information to easily identify 5 main elements: who is important, what is he/she/it doing, how do the actions happen, with what purpose and where and when does the story take place. With the help of BDSi data scientists, a number of requirements are implemented to contribute to the quality of the analysis: splitting into main and subordinate clauses, identifying words that refer to the same person or object across the story and taking negations into account. continue reading...

During social crises like hostage situations, terrorist attacks, and mass shootings, platforms like X are often used by the public and law enforcement to share updates and communicate. However, because this information is publicly accessible, it can also be misused by individuals such as hostage-takers. In this project, social psychologists and BDSI data analysists will collect microblogging messages shared during various social crises. The next step will involve exploring whether content analysis to identify risky messages can be automated. This research aims to enhance police preparedness for managing social crises in the digital world. continue reading...

Narratives or stories in serious games can benefit learning because they facilitate motivation and knowledge construction. Narratives also present an essential element in serious games. Here narratives can vary, from providing an interesting background to playing a crucial interactive role in game completion. Moreover, different types of narratives can be distinguished. The presence and type of a narrative may be an influential factor in game-based learning. The envisioned meta-review seeks to offer a comprehensive overview of the relationship between narratives and digital game-based learning, exploring conditions for effectiveness and presenting a research agenda to address unresolved issues in this field. continue reading...

This grant proposal seeks to address key challenges in the integration of reinforcement learning (RL) into app-based mental health interventions, specifically for the creation of just-in-time adaptive interventions (JITAIs). The main obstacles include developing RL algorithms that don’t require large datasets, are resource-efficient on smartphones, and can balance between trying new intervention strategies and using proven ones. The proposal outlines a collaboration with the Behavioral Data Science Incubator (BDSI) and the BMSLab to overcome these challenges by leveraging the BDSI expertise to implement cutting-edge, innovative RL algorithms suitable for JITAIs within the constraints of existing app platforms. continue reading...

This project is looking to improve the way we can study which mental health treatments work best for each person. Usually, researchers use group studies to test whether treatments work, but these don't always tell us what works for each individual. In this project, a new approach is developed for testing treatments more personally and more often using mobile technology. This will help researchers and clinicians understand what treatment is best for each person and may lead to improved mental health treatments. continue reading...