Typical errors analysis by Russian-speaking students in German texts in the learner corpus pact

In the digitalization era, the attention of foreign language teachers is increasingly turned to linguistic corpora, in particular, learner corpora. The corpora potential and importance in foreign language learning and teaching can hardly be overestimated. The aim of this paper is to present the possibilities of PACT (Petrozavodsk Annotated Corpus of Texts), a Corpus of Student Writings in a foreign language, for analyzing students' typical errors. The authors of the paper present dia-grams based on the corpus, showing the distribution of errors by grade. The graph of students' errors from 1st to 5th year gives a clear idea that by the end of the studies grammatical errors go into the background, but the number of stylistic, logical, discursive and lexical errors increases. The number of lexical, orthographic, punctuation and word order errors exceeds other classes, remaining high throughout the whole study period from the 1st to the 5th year. The reasons for this are largely due to interference with both the native Russian language and English as the first foreign language. The PACT corpus helps teachers in preventing learners' errors, and gives students the opportunity to work independently on their errors and to design an individual educational trajectory.