Extractor Features
Identify and evaluate patterns
Extractor identifies and extracting information "entities" such as names, locations, organizations and so on in the documents and other content that are spread across your business. These entities enrich that content with a structure that can be used to reveal and evaluate high-value patterns. This structure also makes the information in these documents actionable with integrated search tools that are designed to take advantage of extracted entities. These include combinations of Parametric Selection, fielded search and full-text search that uses the extracted entities to enhance relevancy.
Enable category browsing and Relational Taxonomies of unclassified content
The information entities that Extractor identifies enriches unstructured content with structure that can be used to quickly deploy category browsing and Relational Taxonomies in content that has not been previously classified into corporate taxonomies. In effect, each entity becomes a category in a taxonomy that your end-users can browse through to uncover information in the context of related information.
Make any content actionable by business intelligence and other applications
What makes the information stored in databases so valuable is its structure. The well-structured tables and fields in databases make the information in them actionable by ERP, CRM, HRMS, business intelligence and other systems. The information entities identified by Extractor enrich content with the same type of structure so it content can be stored and managed in databases. This directly increases the content's value by making it actionable by other business applications for further evaluation and processing.
Similarly, the structure that Extractor adds to unstructured content can be used to create XML documents. This lets you rapidly repurpose and reuse the content just like you would use any other XML documents.
Enhance search relevance in "entity-intensive" environments
The entities extracted by Extractor can enhance the relevancy of full-text search in "entity-intensive" environments by analyzing text for meaning. For example, in the memos, letters and other content found in patient medical records, Extractor can identify entities such as:
Extractor then tags each letter with this information. So when a hospital administrator searches for a doctor's referral letter for a patient with the name "Jim Smith," the medical records of patients named Jim Smith will be returned, while records of patients who were treated by Dr. Jim Smith will not be.
Rapidly deploy out-of the-box solutions
Extractor provides the 14 most common information entities ready to deploy out-of-the-box. These include:
Extract the information you need to meet specific objectives
Besides the provided entities, Extractor lets you easily build and deploy custom entities to meet your specific objectives. These can include such things as part numbers in manufacturing systems, sell and buy ratings in equity research applications, and the names of suspected terrorist organizations in intelligence environments.
Implement complete, robust solutions
Extractor is the only product on the market that is designed to integrate directly with the search, classification and recommendation features that are required to identify patterns and take action on the extracted entities. This means each component of your solution is optimized to work together with the names, places and other entities extracted from your content.
Scale effortlessly as your content grows and requirements change
Because Extractor is designed to integrate seamlessly with Verity K2's search, classification and recommendation features, it can also scale limitlessly alongside the proven K2 architecture. This both ensures optimal performance and protects your software investment as your content grows, your information environment evolves and your requirements change.
