Top companies trust Datastreamer to integrate, enrich, join, and apply their web data needs.
Detect mentions of people, organizations, locations, filenames, phone numbers, crypto addresses, and more.
Entities are elements of relevance or interest in the text. Tisane extracts both standard entities and those relevant to trust & safety/law enforcement applications.
Standard entities are names of people, their social roles, organizations, places, and so on. We also extract cryptocurrency addresses, bank accounts, credit card numbers, phone numbers, software package names, and more.
Every entity entry is an object made of:
type - the type of the entityname - a standard name, if exists; otherwise, the string that was loggedsubtypes - more detailed additional typessubtype - the first subtype (for backward compatibility purposes)mentions - an array of all detected mentions, with:
offsetlengthsentence_indextextwikidata - a Wikidata ID, if existsOpen Measures collection on OK(Odnoklassniki) began by aggregating around 5,000 OK group URLs that appeared across the existing collections and fed these into the crawler. The crawler then spiders across these groups by collecting posts, comments and group members.
How Datastreamer works
Quickly apply Tisane Entity Extraction to Open Measures Odnoklassniki with a Datstreamer Pipeline.
Web data is the foundation of any pipeline. You can leverage a wide range of data sources to power your pipelines—whether it's data from our partner network, your internal systems, or any publicly available web data.
Datastreamer lets you accelerate your data usage by applying operations like structuring, enriching, joining, and filtering—choose from hundreds of prebuilt, plug-and-play options.
Supercharge your data workflows with Datastreamer. Add flexibility to your Pipelines and put an end to the common bottlenecks in handling web data.
Add Datastreamer components to your data stack and explore its full capabilities
