Sorry! De informatie die je zoekt, is enkel beschikbaar in het Engels.
This programme is saved in My Study Choice.
Something went wrong with processing the request.
Something went wrong with processing the request.

MatchMaking: an app to promote cross-faculty collaboration

At VU many people work on similar topics at different departments or faculties without being aware of their overlap in research topics. Interdisciplinary research can be rewarding as well as challenging. The first step is to become aware of others' interests!

At the Sustainability Office, we're passionate about connecting researchers interested in sustainability. We're developing an app that helps you find peers with similar research interests across faculties. I've attached some screenshots for a preview below.

Our app provides insights into your profile based on thesis topics and shows connections with other researchers. It's a valuable tool for identifying potential research partners within VU who share your interests.

The SDGs created by the United Nations, are designed to outline the primary global priorities for the coming years to enhance sustainability. These goals, while focused on sustainability, encompass a broad spectrum of objectives that extend beyond environmental concerns. As a result, the SDGs hold significance for individuals and organizations beyond those directly engaged in sustainability efforts. The topics next to the SDGs reflect the texts that are collected and could treat topics closely but also more loosely or even not at all related to sustainability

Creating individual profiles based on the SDGs and their associated topics is a practical way to categorize and understand the interests and expertise of different individuals. These profiles can then be used to establish connections between people. When two individuals share common topics in their profiles, it signifies shared interests or expertise, making them suitable for network links.

The full version will be released soon. If you want to join our project, fill in the survey!

Find your match!

Find your match!

Are you curious to find out who is your match? 

Try the app by clicking here

Technical background

The SDGs have been estimated by fine-tuning the ADA model on a selection of texts, labeled by the OSDG community. By selecting only those texts in this public repository that are highly likely to reflect a SDG, we ensured that the model would be able to find patterns in the data that distinguishes these texts from the others. Separate models were fine-tuned for each SDG, which led to an average F1 score of 0.95 (SD = 0.077). The F1 score is a weighted average of those texts that are correctly identified as reflecting the SDG, and those that are NOT reflecting the SDG.

The topics noted next to the SDGs are the topics obtained using Latent Dirichlet Allocation (LDA). LDA is a powerful technique for discovering hidden patterns and themes within a corpus of text. When words are frequently found together in documents, LDA can group them into topics, which helps in understanding the underlying structure of the text data. The best model with the current dataset contained 12 topics. For each model we assigned a label based the most frequent words which you can observe in the profiles. The percentage shows the number of texts that are likely to be clustered in this topic.

Future iterations of this approach could become more sophisticated by considering the strength of these ties. This might involve quantifying the degree of shared interest or expertise and assessing the depth of the connection. Such refinements can lead to more nuanced and valuable network structures that promote collaboration and knowledge sharing among individuals with shared interests.

Would you like to know more?

Contact our Sustainability Officer Research: