Inxight Software, Inc. announced the next level in automatic categorization. Inxight Categorizer 2.0 is an application that efficiently and automatically classifies unstructured data, such as web pages, e-mail, word processing documents, presentations, PDF and text into pre-defined categories to significantly improve and add value to information found on portals and corporate knowledge libraries. Using Xerox PARC and Xerox XRCE advanced language and pattern matching technologies, Categorizer takes classification beyond simple methods such as rule-based systems and pure pattern recognition techniques. Categorizer advances its technology by combining linguistic processing engines with patented mathematic algorithms and helps users to discover context through actual language. For example, the word “ground” can mean ground pepper, the ground that you stand on, or the ground for a wire. Categorizer is able to distinguish between the meaning of the three “grounds” in its context. Another example is the word “teach.” Categorizer identifies and associates words that have the same meaning, depending on how the word is used, like teaches, teaching and taught. Moreover, Categorizer maintains its scalability even as the number of categories or documents grow. New features of Categorizer 2.0 include: Client API coded in Java and C; Database independence (designed for integration to any database or flat file); Added administrative tools to tune accuracy and performance; Support for most file formats including MS Office, Adobe PDF, e-mail, web pages and news feeds; Support for 12 European languages; XML compatibility which enables integration to most enterprise portal applications. Inxight Categorizer is expected to be available in December directly through Inxight corporate sales.