The OECD provides evidence-based advice to countries to foster science, technology and innovation (STI) policies that enhance productivity and support sustainable and inclusive economic growth, strengthen public research institutions and support the creation of science and technology-based firms. Using natural language processing techniques, this project aims to answer the question how new tools and approaches can improve STI policy analysis and public decision making by bringing data scientists together with country delegates. Specifically, participants will be tasked with analysing Twitter data gathered using the Twitter Research API. The goal is to link countries’ STI strategy priorities and instruments to the public discourse on Twitter to help policy makers understand the impact of their policies in domains such as innovation for sustainability and the green transition. Should the team be able to come up with a promising solution, they will have the opportunity to present their findings to an international audience of country delegates and policy experts at an upcoming meeting hosted by the OECD in Paris in December 2022.
Roads and bridges connect people from different places, across rivers and deep valleys. They are fundamental to economic growth and access to basic services. Helvetas helps to build and maintain rural infrastructure that improves access to markets, schools and health centres.
This project focuses on bridges in northern Nepal, where the geography dictates its agrarian population to live in scattered, rural settlements in the mid and upper hills as well as in the high mountains. Over the past decades, with the contribution of Helvetas, many trail bridges have been built, facilitating communication and movement of goods, services and people. The impact of these infrastructure projects has been widely considered to be very positive.
The goal of this project is to use machine learning to analyse satellite imagery (and other datasets such as Open Street Maps) of the surroundings of these bridges, in order to validate these positive changes. Specifically, by detecting trends over time for instance in settlement patterns or access to service centres, the effects of these infrastructure projects could be measured.
The Internal Displacement Monitoring Centre (IDMC) is the world’s definitive source of data and analysis on internal displacement. Their work informs policy and operational decisions that improve the lives of the millions of people living in internal displacement. Besides national governments, local authorities, UN organisations, etc. IDMC relies on media monitoring to gather information on internal displacement. For this purpose, media sources play a significant role, particularly when governments lack the capacity or will to collect data.
The goal of this project is to help IDMC develop a better media monitoring tool, replacing the current black box approach with a transparent model which has functionalities that better reflect the needs of IDMC’s monitoring experts. The Hack4Good team will work closely together with technical experts from IDMC and is challenged to write reproducible and well-documented code to ensure the tool will easily be picked up. The impact of this project is that IDMC increases the scope of their monitoring and better identifies internal displacement occurrences.
Nitrogen emissions cause eutrophication and soil acidification and damage sensitive ecosystems such as forests, dry meadows, or peatlands. Agriculture and animal husbandry in particular is the main source of nitrogen emissions, which threaten biodiversity and weaken the resilience of ecosystems.
Construction projects on farms (e.g. a planned new stable), may have an effect on ammonia emissions. Thus, WWF regularly controls whether construction projects respect the Swiss environmental law. If not, they file a legal complaint. Due to the large number of building project applications, the vast amount of publication sources and the limited personnel resources, a systematic preliminary examination of building applications is not possible.
The Hack4Good team will be challenged to create an automated analysis pipeline to identify building permits with potentially adverse impacts on biodiversity. This project aims to improve the control of cantonal permit authorities and thus to enforce environmental regulations in the building permit procedure. As a result, fewer building applications that lead to a deterioration of the environment will be approved.
Every year, about one-third of the total food produced for human consumption is wasted.
The lack of adequate facilities where the produce can be stored post-harvest plays a key
role, especially in developing economies in Africa, Asia, and Latin America. BASE (the Basel
Agency for Sustainable Energy) and Empa have developed a mobile application for local entrepreneurs who offer solar-powered cold rooms. Farmers can request capacities in the cold rooms through the app. The app also allows farmers to make informed decisions about when and where to sell the produce stored in the cold room to reduce post-harvest loss and maximise their income. The app collects data about cold room usage as well as farmers’ data on crop production, post-harvest loss, and selling prices.
The task of the Hack4Good student team will be to create a pipeline for automated app data
extraction and analysis. BASE aims to apply this data to:
- evaluate the project’s impact
- create a machine learning model that predicts the benefit of cold room usage in terms of post-harvest loss reduction and income increase for prospective users.
To provide adequate humanitarian aid at the right place and time it is critical to understand the situation and needs of the population in conflict zones and inaccessible areas. IMPACT Initiatives conducts regular, extensive surveys in 18 countries to assess the humanitarian situation. The direct control over the work of those who physically collect the data (called enumerators) is very limited, and several falsification cases have been detected such as fake answers or random inputs. IMPACT Initiatives thus have developed data cleaning standards and processes. These require a lot of human intervention and are quite complex. This project aims to support the detection of fake survey answers with machine learning or statistical methods. Extensive past survey data are available. The goals of this project are to establish a baseline for enumerator behaviour, identify unusual data points and, optionally, package the developed method into a tool to make it accessible to all of IMPACT’s teams. As a result, IMPACT will have more accurate data to inform its actions in the field.