Auto-classification is a term that will scan the contents of a document and automatically assign categories and keywords found in the document contents. But auto-classification systems are not one size fits all. Some require training sets and if using rule building, multiple iterations are needed, and rules maintained and tested to improve accuracy. Others require outside application specialists or learning new languages. Preconfigured systems may be tailored to your industry but not your corpus of content.
The Concept Searching Difference
Concept based auto-classification
Concept Searching’s core technology supporting our auto-classification capability is unique in the industry. Auto-classification clues are built by automatically generating multi-word concepts from clients’ own content. The classification engine then classifies the content to one or more nodes in one or more taxonomies inserting the metadata either into the managed metadata fields in SharePoint or directly into the document properties, in the case of content on the file shares. Elimination of complicated Boolean expressions, proximity, or scripting, the metadata is consistent and reflects the terminology without having to customize the output. Since the metadata is very precise, the auto-classification engine will identify intelligent content in context that can be used in applications such as records management, identifying confidential data, migration, text analytics, content management, secure collaboration, compliance and high performance search applications.