This paper describes the first prototype of an automated tool for detecting collocation errors in texts written by non-native speakers of English. Candidate strings are extracted by pattern matching over POS-tagged text. Since learner texts often contain spelling and morphological errors, the tool attempts to automatically correct them in order to…

Describes three pilot projects carried out with a special version of a computer-assisted language-learning activity, known as SEQUITUR, that seeks to develop awareness of cohesive devices and coherence features by displaying the start of a text and offering possible continuations. Usage logs of responses given by native and nonnative speakers of…