Title: Experiments in computer-assisted scoring of content-oriented constructed response items Abstract: In educational testing, a constructed response (CR) item is a type of open-ended question or prompt which elicits free production, rather than specifying a closed set of answers from which a test-taker selects (as in, for instance, multiple-choice tests). Due to their open-ended nature, CRs need to be graded by human readers, which makes the overall process costly and time-consuming. Not surprisingly then, the task of computer-based scoring of this type of items is an interesting application of NLP. In this presentation i will talk about /work-in-progress/ on computer-/assisted/ meaning-oriented scoring of CRs and prospects for streamlining the grading task via interactive supervision.