From planning to beta – how we are building a tool for investigators and data analysts
In response to the growing needs of investigative services, prosecutors and digital-data analysts, at FireNet we have begun work on an innovative tool for analysing very large volumes of e-mail. It will be an advanced platform supporting the investigative process, enabling the searching, filtering, sorting and categorising of messages – even in huge data sets that until now have been difficult to process effectively.
Project goal
The tool's aim is to create an intuitive and fast analytical and presentation environment that will allow you to:
- quickly search through hundreds of thousands of e-mails,
- filter data by sender, recipient, date, keywords and attachments,
- categorise messages by subject or event type,
- identify potential connections and communication patterns,
- build data sets for use in machine learning and AI.
The planning stage – practitioners' needs
The project began with conversations with officers, analysts and digital-forensics experts who deal with enormous e-mail archives in their daily work. We identified specific problems:
- time-consuming searching of messages,
- a lack of automation in identifying relevant e-mails,
- the need to quickly create reports and summaries,
- the lack of tools for preparing data for further AI analysis.
The development stage – from code to the first tests
In the next stage, FireNet's team of developers and analysts began work on the modules:
- Message-processing engine – handling large files (e.g. PST, MBOX) and decoding content,
- Presentation layer – a modern interface that makes navigation and investigative work easier,
- Search and filtering module – a fast and configurable query system,
- AI integration – preparing labelled data for training machine-learning models,
- Data export and reporting – generating clear results for further use.
Beta version – the first tests in an investigative environment
The tool is currently in a closed beta, tested internally and at selected partner institutions. The first results are very promising – in testing we managed to search, sort and extract key messages from an archive of more than 1,000,000 e-mails in under 20 minutes.
What next?
In the coming months we plan to:
- continue testing and gathering user feedback,
- implement additional features (e.g. graph analysis of connections between senders),
- expand the export and reporting modules,
- prepare a production version for institutional clients.
The tool is being created with real investigative and analytical needs in mind. By combining the knowledge of our specialists with the experience of investigative services, we are building a solution that will streamline work where time, accuracy and reliability matter.
Want to learn more? Get in touch – we invite public institutions and organisations interested in testing the tool to work with us.