Hacking for a good cause
During an 8-week programme, data science talents from ETH Zurich solved data-related problems for non-profit organisations that support social causes.
external pageHack4Good is a student-led initiative, organised by the external pageAnalytics Club, that matches students with an interest in data science with non-profit organisations working on social causes or environmental issues. By analysing and processing data, the students help a non-profit organisation to increase the positive benefits ("impact") of their operations.
As part of the 5th edition of their annual programme, 25 participants worked in autumn 2022 during eight weeks in six different teams on solving specific data-related problems of a total of six non-profit organisations (NPOs). As NPOs often lack the necessary resources to carry out data-based analyses, the participants of Hack4Good offer a valuable service to NPOs - on a voluntary basis!
Innovative implementation of data analysis methods in various application areas
One team of students helped the external pageInternal Displacement Monitoring Centre (IDMC) to automatically collect information on internal migration in real time from media reports. A database of more than 8,000 media reports in 100 different languages was scanned daily and Natural Language Processing was used to classify the reason for migration and capture the number of people affected.
For the think-and-do tank external pageIMPACT, which carries out humanitarian monitoring and evaluation activities, a team created an algorithm to detect fake survey results in polls. One of the objectives was to detect and prevent fraudulent completion of survey forms by enumerators. The algorithm flags possibly fraud survey forms, which could then be checked for authenticity by IMPACT staff.
“Really impressive work by the students working on this challenge! Pending further testing, the solution they have designed has a lot of potential to help strengthen quality of data collected across different conflict settings.”Olivier Cecchi, IMPACT
One project also directly related to Switzerland. For the WWF, a team developed a programme that checks applications for agricultural building permits for their impact on biodiversity, with a focus on nitrogen oxide emissions. For this purpose, an algorithm was developed that extracted the location of the construction project from the application forms, which differ from canton to canton, and checked its distance from zones with sensitive ecosystems. This enables WWF to quickly identify and object to construction projects with a potential negative impact on biodiversity.
“Thanks to Hack4Good, we were able to pilot an early-stage data-driven idea, in order to support our cantonal sections in their daily work.”Attila Steinegger, WWF
The descriptions of three further projects, which were conducted for the NPOs Base, Helvetas and OECD, can be found on the external pageHack4Good website.
A win-win situation for all involved
In the Hack4Good programme, students learn to combine their programming and data-analytical skills with applied and collaborative work in teams that are very diverse regarding study programme, origin, and gender. In doing so, they can also network with actors in the humanitarian sector and thus gain experience in transdisciplinary cooperation between science and practice. For their part, the NPOs benefit from the creativity and fresh perspectives of a young, committed generation of future ETH graduates. The data analysis tools developed by the students are all designed with a user-friendly interface and are now being distributed to the organisations for their own purposes to increase the impact of their work.
The Hack4Good organising committee is now recruiting a new group of students for the 6th edition of Hack4Good in autumn 2023. It would be interesting if ETH itself could provide a data set, for example to better understand the impact of its climate protection measures and to further promote ETH Zurich's path towards the net zero goal.
“It was really nice to get practical experience with all the steps of a data science pipeline, while knowing that we were doing meaningful work.”Jacob (participant)
Further informationen
Always up to date
Would you like to always receive the most important internal information and news from ETH Zurich? Then subscribe to the "internal news" newsletter and visit Staffnet, the information portal for ETH employees.