A statistically based program has been written which learns to distinguish between languages. The amount of training text that such a program needs is surprisingly small, and the amount of text needed to make an identification is also quite small. The program incorporates no linguistic presuppositions other than the assumption that text can be encoded as a string of bytes. Such a program can be used to determine which language small bits of text are in. It also shows a potential for what might be called `statistical philology' in that it may be applied directly to phonetic transcriptions to help elucidate family trees among language dialects. A variant of this program has been shown to be useful as a quality control in biochemistry. In this application, genetic sequences are assumed to be expressions in a language peculiar to the organism from which the sequence is taken. Thus language identification becomes species identification.

Resumen: En este artículo se comparan tres sistemas estadísticos de identificación de idioma. Se presenta también un estudio detallado de la influencia de algunos factores importantes sobre la precisión de los sistemas. Estos factores son: la medida del corpus de entrenamiento, la cantidad de texto que se quiere clasificar y las lenguas entre las cuales el sistema es capaz de distinguir (se estudiará tanto el número de lenguas cómo cuáles son esas lenguas). Palabras clave: identificación de idioma, sistemas estadísticos, multilinguismo, modelos de Markov visibles, vectores de frecuencias de trigramas, categorización de textos basada en-gramas Abstract: In this work three different statistical language identification methods are compared, and a detailed study of the influence on those systems of some basic parameters is performed. The analyzed parameters are the size of the train set, the amount of text that we want to classify and the languages the system is able to distinguish (it will be studied not only the influence of the number of languages but also the influence of wich are the considered languages).