Configure the Phrase Classification, Table Analysis and Format Analysis engines using phrases or strings in the desired non-western language on the corresponding non-western PC or on any PC where this language is the default system’s Unicode language.
This is also applicable for all other BIC engines that do not require any visual input in non-native languages like the Brainware Extraction engine, the Brainware and ASSA Classification engines, the Brainware Layout Classification engine.
While the support option for non-western languages always looks for a one-to-one character transformation, the CJKT support may lead to an increase in text length of the transliterated object. The multi-character representation is phonetically more precise but increases the original text length. The attributes of the SCBCroWorktext objects like TextLength refer to the original character length. The attribute TextProperty and the Len method return the length of the transliterated text. This limitation applies only during the "transliteration phase", which usually corresponds to internal core classification and extraction and also involves some script events, like PostEvaluate, executed during the core extraction.
To use a language other than English, change the region and language settings of your system.
BIC provides a generic approach to the support of “non-western” languages, where essential parts of the alphabet are not covered by letters of the English alphabet with the same key codes.
This includes the following languages.
- Greek
- Polish
- Russian
- Chinese, Japanese, Korean, and Taiwanese (CJKT)
Designer applies an automatic phonetic translation from non-western languages into English and then an unambiguous conversion back to the original language. The conversion applies for the following workflow steps in BIC.
- Classification
- Extraction
- Validation
- Learning
You can execute these processes on a non-native PC, such as a United Kingdom (UK) or United States (US) workstation or server, to process documents that contain auxiliary information and information that is extracted in supported non-western languages.
The following engines allow you to enter phrases and format strings with visual input using the characters of the currently supported non-western languages.
- Phrase Classification engine
- Table Analysis engine
- Format Analysis engine
- Associative Search engine. This engine includes the ability to import Unicode CSV files.