Systemic Relational Insight

In today's interconnected world, it's easy to stumble into complexity: relating seemingly simple problems in space and community quickly leads us into a dimension where applying formulas or best practices is not enough. The decisions we make are influenced by myriad factors, and often we don't know how these factors relate to each other, scattered across heterogeneous data types (qualitative and quantitative).
We increasingly talk about data without considering the context in which it is collected, and this leads us to make decisions based on information that doesn't represent reality, let alone the community and territory.
Traditional methods of data analysis and decision support often fail to address these complexities. This is where Systemic Relational Insights (SRI) come into play.

SRIs are a new method and sense-making tool that uses a chain of algorithms to generate abductive explanations to facilitate problem understanding and interaction between qualitative data collected from interaction with people, quantitative data, and scientific literature. We consider them a vehicle for hybrid collective intelligence, combining human input and supervision with algorithmic support. It is a transparent, secure, and traceable approach that helps citizens and administrators establish solid shared foundations for better decisions.

SRIs are a description of a possible relationship between two or more phenomena, resulting from the synthesis of experience, knowledge, and data, validated by a community representative of the system itself.

They are a powerful tool for analyzing and understanding socio-technical-natural systems, supporting participatory and informed decision-making processes, promoting social innovation and co-design of solutions, and creating a shared and evolving knowledge base.

Analysis map of a problem in the Spilamberto (MO) community, with the emergence of some SRIs

Distinctive Features

SRIs differ from traditional data analysis methods in several key aspects:

• They use a hybrid human-machine approach, combining human input with algorithmic support. In fact, they are best used only when the algorithm's output is validated by a sufficient number of people.
• They focus on the relational dimension, reorganizing input data to give more weight to important things and concepts expressed by people, rather than the generalist trends of models trained on data we don't know.
• They integrate heterogeneous data, including qualitative (for example, tables and smart sensor data) and quantitative data, collected during participatory processes, relating them also to scientific literature corpus, thus overlapping different ways of understanding the world and producing knowledge.
• They are validated by the community itself, ensuring greater relevance and acceptance of results.

Where to Find Them

Systemic Relational Insights are being integrated into a web app currently under development. After an initial validation of the interaction experience, they will also be implemented as APIs for their use within your applications and third-party platforms for managing civic engagement and processes.