Intelligent analysis of complaints and suggestions in Valencia using data intelligence: transforming citizens' voices into data-driven urban decisions
Public administrations generate an enormous amount of data every day. However, having access to this data does not necessarily mean possessing knowledge. In many cases, the information is fragmented, in different formats, with quality issues or with a level of complexity that makes it difficult to actually use for decision-making.
This is the context for the use case developed by Andrea Bonetti, a member of our research team at IDAL (Intelligent Data Analysis Laboratory) at the University of Valencia. His proposal consists of a data intelligence-based tool capable of transforming citizen complaints and suggestions—extracted from the Valencia City Council's Open Data Portal— into useful and actionable knowledge for urban management.
More specifically, it was developed as his Master's Thesis (Master's Degree in Data Science from the University of Valencia) and is also aligned with the European Citcom.ai project. It is conceived as the basis for a scalable solution in the future if the appropriate data is available, being adaptable to different municipal departments and even to other cities interested in listening to and analysing the voice of their citizens in a structured way.
From isolated data to actionable knowledge
The initial problem, which led to the selection of this idea as the basis for his work, was clear: the existence of more than 70,000 records of citizen complaints and suggestions accumulated since 2020, categorised by topics and subtopics, linked to different neighbourhoods and districts of the city, but without an integrated system that would allow spatial, temporal or thematic patterns to be extracted intuitively.
In addition, there were several significant technical challenges, such as:
- The existence of geospatial data for Valencia's 88 neighbourhoods in GeoJSON format.
- The need to clean and purge records with incomplete information.
- The standardisation of bilingual neighbourhood names using fuzzy matching techniques to ensure consistency and reproducibility in the analysis.
- The integration of auxiliary context data, such as school routes, hospitals, senior centres, and metro and tram stations.
The final result, following Andrea's work and the resolution of all technical difficulties, was the development of an interactive dashboard that combines data analysis, geospatial visualisation and machine learning techniques to provide a comprehensive and unified view of citizen incidents. Here you can see a video of the tool.
The tool is based on a clear conceptual approach: understanding citizens as a sensor distributed throughout urban territory. Each complaint or suggestion is not just an isolated incident, but a signal that can reveal structural patterns in certain neighbourhoods, recurring problems in certain services or inequalities in participation channels.
The dashboard developed allows users to:
- View the geographical distribution of complaints by neighbourhood.
- Analyse the temporal evolution of the most recurring issues.
- Detect possible biases in the input channels (web, telephone, electronic headquarters, etc.).
- Identify areas where the citizen communication channel could be underused or deteriorated
Furthermore, one of the most innovative features is its no-code nature: the tool is designed so that municipal technicians can explore information and make informed decisions without the need for advanced knowledge of programming or data science. This reduces technical barriers and facilitates the actual incorporation of analytics into public management.
Scalability and future prospects
The system's modular architecture allows its functionalities to be easily expanded. New layers of data can be incorporated, real-time information can be integrated, or the model can be adapted to other urban contexts simply by adding new modules to the main panel.
This use case demonstrates how open data can move beyond being mere information repositories to become strategic governance tools. When advanced analysis techniques, geospatial visualisation and artificial intelligence are combined, the voice of citizens ceases to be scattered noise and becomes a structured source of knowledge.
In short, the work carried out by our colleague Andrea within the framework of Citcom.ai demonstrates that it is possible to move towards a more intelligent, proactive and data-driven administration, capable of anticipating problems, identifying territorial inequalities and strengthening institutional trust through active and systematic listening.
If you found this use case interesting and would like to explore how to apply a similar solution in your municipality or organisation, please do not hesitate to contact us through the Contact section of our website. We will be happy to help you turn your data into decisions!
The authors would like to thank the Regional Ministry of Innovation, Industry, Trade and Tourism (Conselleria d'Innovació, Indústria, Comerç i Turisme de la Generalitat Valenciana) for its funding under Grant Agreement No. 101100728, as part of the European Commission’s Citcom.ai project.