Transformer-Based Real-Word Spelling Error Feedback with Configurable Confusion Sets

Real-word spelling errors (RWSEs) pose special challenges for detection methods, as they 'hide' in the form of another existing word and in many cases even fit in syntactically.We present a modern Transformer-based implementation of earlier probabilistic methods based on confusion sets and show that RWSEs can be detected with a good balance between missing errors and raising too many false alarms.The confusion sets are dynamically configurable, allowing teachers to easily adjust which errors trigger feedback.1.The name comes from the Greek work [word] for sun.2. These plants are more tolerance [tolerant] to drought.People with lots of honey usually live in big houses.a word is replaced with a mask token and the library returns the most likely fillers and their probabilities.So for the resulting masked sentence People with lots of [MASK] usually live in big houses.