Most IT systems cannot hold product data in a form of classified tables of attributes and values. Their main method for holding such data is by free text description(s). In order to allow basic search and compare activities with the product descriptions, within the IT system, the description should be standardized, consistent and in the relevant language. Product description should conform to a variety of constraints, depending on the target IT system e.g., maximum length of a description.

The challenge is to provide the user with a tool that will enable to produce standard, consistent descriptions from the tabulated product data, in any desired language, conformed to the given constraints.

The roll of the Descriptor is to automatically produce variety of standard, multi-lingual descriptions under a set of constraints.

We use the Descriptor after extraction and de-duplication. The user defines templates for each category to set the description length, order of attributes, language, lexicons and other parameters.

The deliverable of the Descriptor is a set of standardized product descriptions for each product.