Most companies already have adequate solutions  for accessing structured information, as it is, for example, stored in databases. However, frequently a lot of important information is only available in an unstructured form, for example, as emails, text documents, Web sites, news feeds, and so on. Up till now, it was unfortunately impossible to systematically query this information.

The Semantic Content Analytics (SCA) solution, developed together with IBM, enables the efficient use of unstructured content and the merging of this content with the structured content at semantic level.

• The analysis of a wide variety of texts: news feeds, documents, emails, Wikis, databases, and so on
• The improvement of text analysis using knowledge models (ontologies)
• The extraction of facts, automatic tagging/classification of documents
• The structured storage of facts in a knowledge model
• Linking to facts from other sources, for example, structured facts from databases, and so on
• The logical derivation of new information
• The targeted querying and graphical processing of all information

 
Discover, save and analyze knowledge

SCA is a solution developed together with IBM for speeding up content analytics processes conducted in companies. SCA can be used to automatically analyze a large amount of unstructured data which is frequently stored in text form. In the process, concrete information, such as, for example, "Company A bought company B", "Company C brought product D onto the market", is extracted and saved in a knowledge base. This information can be directly accessed using the knowledge base.

The knowledge base itself can be used to derive new information from the existing data. This takes place using rules that are contained in the knowledge base and evaluated by SCA.This makes it possible to make knowledge available that is implicitly visible. For example, it can be used to show all of the products from a parent company that are offered by one of the subsidiaries. This information is not explicitly available in the analyzed texts but it is derived from the existing information using rules.