Use the same documents that you intend to categorize. It is preferable to choose rich content model document.

Intellexer Categorizer processes documents of the following file formats:

• TXT (Plain text)
• HTML (Hypertext Markup Language)
• DOC (Microsoft Word)
• PPT (Microsoft PowerPoint)
• RTF (Rich Text Format)
• CHM (Windows Help file)
• PDF (Portable Document Format)
• DOCX (Microsoft Word 2007)
• MHT (Multipurpose Internet Mail Extension HTML)

There are some limitations in the current version of Intellexer Categorizer. They are:

1. You can create up to 30 categories for one project;
2. Maximum amount of documents in a single project cannot exceed 300

If you have more than 300 documents in one folder, they can be easily processed in several steps:

1. Create categories and assign model documents
2. Add the first 300 documents to the project
3. Distribute processed documents to other folders after categorization using the command “Move Documents”
4. Clean the field Documents and add the next 300 documents
5. Repeat these steps until you distribute all documents in the original folder

Take one or several documents which you believe were categorized incorrectly and add them as model documents to the category you want to associate them with.

You may set any threshold of Proximity (relevance) in menu Category \ Relevance Settings

This is a conditional parameter, calculated by the semantic core of the program. Proximity shows how one document is similar within a meaning to another document or to a group of documents. For the convenience of clients this parameter is calculated in %.

It should contain at least one model document. The optimal quantity of documents for certain categorization varies from 8 to 10 different documents of diverse subjects.

This is a common document, which will be used as an example for finding similar documents. A feature of our program allows you to use several model documents for one category.