COVID-19 Digital Campus

Wellbeing

With most of our teaching activities taking place online, we miss out on normal interaction with our students and thus also pick up fewer signals about their well-being. How are they actually doing in this tough situation; how are they really feeling? This requires extra attention. First-year students in particular can have a difficult time of it. They don't have any friends in Delft yet, and haven’t established ties with their classmates or lecturers. Using new tools we can gain more insight into the mental well-being of our students and we can also do something about improving it.

Dandelion

Conversational Crowd Computing Platform

Conversational agent Dandelion is a feedback tool for students. With most of our teaching activities taking place online, we miss out on normal interaction with our students and thus also pick up fewer signals about their well-being. How are they actually doing in this tough situation; how are they really feeling? This requires extra attention. 

The conversational agent is a tool to get feedback from students as non-invasively as possible. Instead of lengthy questionnaires, the system uses a chat interface to ask simple questions –whether they felt safe or what their state of mind is, for example. Privacy is ensured as the answers are not linked to individuals. Additionally, students can also share information, for example photos of areas that are too crowded. Such information can then be used in the Campus Mobility Dashboard.

Trainbot

A training tool for stress management

COVID-19 leads to stress and loneliness among students. First-year students in particular can have a difficult time of it. They don't have any friends in Delft yet and haven’t established ties with their classmates or lecturers. This is why there is more focus on mentor programmes, where senior students mentor first year students. Besides answering practical questions about studying at the university, these mentors will also have to pay more attention to students’ wellbeing.

Trainbot is a conversational interface that trains non-experts the technique of Motivational Interviewing, a counselling approach that is used to treat anxiety, depression, and other mental problems.

Trainbot was developed in collaboration with TU Eindhoven. Results from earlier controlled experiments have shown that workers using Trainbot:

  • felt lesser pressure than those in the control group
  • provided psychological interventions that were rated consistently higher by psychologists than those in the control group
  • felt a higher self-efficacy in helping deal with stress management after the training process

We would therefore expect our student mentors to benefit from learning how to provide support to stressed students, and when to escalate cases where professional help is required. We also expect that students with mentors who have completed training with Trainbot, will be more satisfied with the interaction with their mentors. To test these expectations, and gain insight into the effectiveness of Trainbot at TU Delft, we will run randomised controlled experiments where half of the student mentors at a given faculty will receive training through Trainbot, and the other half serving as the control group will not.

Building rhythms

Data as a means of discussion

The Building Rhythms project investigates how privacy-preserving sensing and public data visualisations can become a platform for understanding COVID-19's impact on campus life, and for supporting health and wellbeing. 

Data as a tool for discussion

We will use anonymous information from sensors in and around buildings to expose daily and weekly patterns in the use of buildings – the buildings’ rhythms. We will then engage staff and students in discussions, so we can jointly try to formulate acceptable interventions to keep campus life safe and to keep student experiences positive. By encouraging students and staff to critically reflect on health risks, personal behaviours and the ethics of ubiquitous data collection and use, the project becomes a learning environment for developing novel approaches for data science and data-centric design education.

Building Rhythms involves:

  • privacy-preserving sensing
  • privacy-aware data processing
  • data visualisation methods
  • data-centric design
  • data ethics

Building Rhythms is a collaboration between the Faculties of Industrial Design Engineering (IDE), Electrical Engineering, Mathematics and Computer Science (EEMCS), and Architecture and the Built Environment (BK).

/* */