Machine Learning
What is machine learning?
Machine learning (ML) is a set of methods for training computers to learn from data, where “learning” generally amounts to the detection of patterns or structures in data. If designed effectively and paired with large amounts of useful data, machines are able to use those patterns to make predictions about the future – all with minimal pre-programmed rules.
can we use it to predict shifts in civic space?
As machine learning progresses and digital data expands, it is increasingly possible to test the key drivers of civic space closure by processing global information at unprecedented speed and depth. A ML model could potentially predict the closure of civic spaces more effectively than traditional approaches. There are several measures of civic space already in existence, but their coverage varies, they are typically annual and released with significant delay, and they are not designed to be predictive. Thus, existing measures are not actionable when closures can happen over weeks and months rather than years.
Forecasting
INSPIRES will test the proposition that machine learning can help identify early flags that civic space may shift and generate opportunities to evaluate the success of interventions that strive to build civil society resilience to potential shocks. The Consortium will create an unprecedented data set to analyze the drivers of shifts in civic space, combining locally-created and curated data sets (see additional information on our Expert Survey page) with extensive digital data sets and longstanding research. DevLab@Duke will use this data to build a forecasting model that will test critical hypotheses around civic space drivers and seek to generate opportunities to understand changes in civic space and enable preventive interventions. INSPIRES will disseminate findings and support stakeholders in understanding, analyzing, and applying machine learning analysis.