Get structured data from any text source using Rosaia's system for data extraction
Fine-grained data extraction at scale
Automate data extraction from unstructured sources
Get a grip on latent data
Extract structured data from any source, regardless of how messy it is.
Reliable results
Get realiable output using a systematic and verifiable process.
Focus on your own expertise
Use your own professional knowledge to focus SEED on data relevant for your profession.
Scale up
Get data from 100, 1.000, 10.000 or more documents; SEED scales too.
Technology and process in one
The inner workings of SEED
Identify relevant texts
SEED queries sources for texts from which structured data may be extracted.
- Queries sources for texts with relevant information
- Processes and splits up texts into atomic fragments
- Sets up a database based on fragments and their origins
Categorize and group texts
SEED categorizes fragments following instructions and uses these categories to group related texts.
- Categorizes fragments based on contents following user instruction
- Connects fragments with the same identifying information
- Groups all fragments using connections between fragments containing identyfing information
Analyse texts in context
SEED creates per group and category a structured summary of all information contained by grouped fragments.
- Analyses information in fragments based on user instruction
- Generates unambiguous summaries of analyzed information
- Structures summaries in JSON, as simply or complex as needed
Convert analyses in data
SEED combines all information per group to develop an archivable object.
- Combines fine-grained data in a organized data object
- Checks for the adherence to prescribed data schemes
- Stores results in a relational database