Language Identification Using Global Statistics of Natural Languages

This article is about a new method which makes it possible to identify the
language of a written document. The method is based on the analysis of simple descriptive
statistics of the given text. These simple statistical features include things like average word
length or consonant congestion.
In order to measure the effectiveness of the method an application has been developed
which can classify English, Hungarian, German, Spanish, Croatian, French and
Norwegian documents by analysing the average word length, the ratio of certain
characters, word endings and consonant congestion.