KMS Suite Products - KITE
Knowledge Identification and Transformation Engine
CSW’s Knowledge Identification and Transformation Engine (KITE) converts corporate structured documents into explicitly structured XML content. It uses a set of custom defined rules to mass convert documents, then validate and store results in an XML content management system.
Document reasoning through rules
The reasoning approach is the fundamental method used by KITE to identify and convert content into knowledge. By applying a set of logical constraints and rules that match the textual content, KITE can generate a much more meaningful and structured end product, reducing the usual time required by more traditional approaches to tidy up the end data.
Most documentation that adheres to even a basic specification will have certain common features that help to define its structure, whether they are a list of common titles that are used throughout, or a specific font weight and size for section headings. Merging these styling rules with a set of business rules and constraints allows KITE to generate better structured XML than purely a style-matching technique. This rules-based processing sets KITE apart from the competition.
Rules
KITE starts with a simple set of customisable rules which are applied to the original document. It then matches words, phrases or reusable components such as product names and can be customised to provide either an exact match, or the highest ranking alternative.
KITE is fully extensible so that it can meet the needs of more complex matching rules or specific document requirements. Separate rules can be applied to different sets of documentation. This allows an archive of rule templates to be built which can be applied to specific instances as required. As more examples pass through KITE, the rules can become more specific and well-defined to produce structure rich data.
KITE - Click to view larger
KITE - Click to download image
Validation
Once KITE has been run across the source document set, a report is generated for user validation. This report contains a list of the matches and suggestions found during the initial pass based on the application rules. The end user decides to accept or ignore each item before running the application again. This approach has the benefit of enabling the user to decide whether any adaptation of the rules would produce better results.
Application
Once the validation report is accepted, the results are then applied to the full source document set. KITE then outputs a fully structured XML document and stores the content directly in an XML Content Management System such as CSW’s Knowledge Engineering System (KES).
Features
- Converts documents from Adobe PDF, Microsoft Word (97-2007), OpenOffice and Corel WordPerfect
- Generates standard XML output
- Simple XML-based rules language, using industry-standard XPath 2
- Built-in web-based rules editor
- Integrates with CSW’s Knowledge Engineering System (KES)
Benefits
- Powerful real-time reasoning platform for execution of custom rules
- Handles semi-structured source documents
- Standards-based integration
- A proven, stable deployment platform
- Rapid application development
For further information on KMS, visit the product website at:
www.xmlkms.co.uk
