Text analysis in unsupported languages

A computer-implemented method for delivering personality-appropriate advertising to a user in an unsupported language, comprising:

• determining one or more global analysis parameters based on backtranslation of a first corpus between supported languages by determining a usage change characteristic from a change frequency across one or more categories between the first corpus and the backtranslated first corpus;
• determining a new personality estimation model for an unsupported language based on the one or more global analysis parameters and a text analysis model for a first supported language by backtranslating a second corpus between the unsupported language and the first supported language and determining a usage change characteristic for the unsupported language based on the backtranslation of the second corpus;
• and
• estimating a personality characteristic of an author of an input text in the unsupported language with the new personality estimation model, using a processor, by counting a number of words from the input text associated with each of a plurality of personality categories;
• and
• advertising a product to a user in the unsupported language based on the estimated personality characteristic.

A text analysis system, comprising:

• a supported language module configured to determine one or more global analysis parameters based on backtranslation of a first corpus between supported languages;
• a language adaptation module configured to determine a new text analysis model for an unsupported language based on the one or more global analysis parameters and a text analysis model for a first supported language;
• and
• a text analysis module comprising a processor configured to analyze an input text in the unsupported language with the new text analysis model using a processor.