Automated Content Enrichment
Use the transformational power of semantic tagging to enrich your content automatically. Our offerings enable you to evaluate different text mining offerings, make a build-or-buy decision, aggregate different services and cherry-pick their best sides to serve your use case in the most optimal way.
How you’ll benefit
Facilitate advanced search capabilities, content organization and non-linear user exploration by improving metadata coverage and consistency in your content
Develop new data products from your unstructured assets, enriched through information extraction, enabling your transformation from a content vendor to a data vendor
Optimize your operational costs and editorial efforts by automating the metadata generation.
Some of the challenges lurking in any natural language content repositories:
Ambiguity
The language is inherently ambiguous. The same word has different meanings in different contexts. For example, Paris, France is a location; Paris Hilton is a person; Paris is a Greek myth hero; Paris, Texas is a movie (and location), etc.
Inconsistency
The same term is frequently used for describing (slightly) different phenomena. This is the case even in a highly formal context (i.e., legal, medical) where there are strict and more popular terms, describing a certain concept.
Variability
Often there are slight variations in naming conventions – i.e., William, Bill, Billy, etc.
Volume
The typical task is to process large bodies of text created by different authors, organizations, cultures and targeted towards different audiences and objectives.
How it works
- Create a knowledge organization system based on a taxonomy or an ontology describing your domain, business or use case.
- Apply automatic tagging to generate rich metadata driven by your model.
- Multiply that by leveraging Natural Language Processing integrated with big knowledge graphs to extract business concepts like people, organizations and locations and the relationships between them.
- Aggregate multiple text mining services and pick the best output from each one of them to compose content enrichment that serves your use case best.
- Use inference to enrich the resulting metadata and provide search results that are not explicitly stated in your content.
Showcase:
News on the Web (NOW)
A free public service, showcasing the opportunities open up before media and publishing companies. Get a real feel of the world where semantic technologies are already shaping the way we search, discover and consume content.
