Spelling correction for an intelligent tutoring system
Our spelling correction program is part of a system for understanding illformed input in an intelligent tutoring system for medical students. Speed and user-friendliness were the most important considerations in the design. The system can correct most kinds of spelling errors including order reversal, missing characters, added characters, and character substitutions. It also handles novel abbreviations and word boundary errors. It is implemented on a Xerox 1108 AI machine in Interlisp-D. The lexicon is stored in a trie structure to speed up searching. We provide the students with a full-screen editor and an input tracer to simplify the input process as much as possible.
KeywordsLexical Entry Disagreement Threshold Intelligent Tutoring System Input String Spelling Error
Unable to display preview. Download preview PDF.
- A.A. Rovick and J.A. Michael. CIRCSIM: An IBM PC Computer Teaching Exercise on Blood Pressure Regulation. XXX IUPS Congress, Vancouver. 1986.Google Scholar
- Y. Zhang, M. Evens, A.A. Rovick, and J.A. Michael. Knowledge Compiler for an Expert Physiology Tutor. Proceedings of ESD/SMI Conference on Expert Systems Dearborn. 1987: 245–259.Google Scholar
- R. Grishman and R. Kittredge. Analyzing Language in Restricted Domains: Sublanguage Description and Processing. Lawrence Erlbaum Associates, Hillsdale, NJ. 1986.Google Scholar
- N. Sager, Sublanguage: Linguistic Phenomenon, Computational Tool. In R. Grishman and R. Kittredge (eds.), 1986: 1–18.Google Scholar
- E. Horowitz and S. Sahni. Fundamentals of Data Structures. Computer Science Press, Rockville, MD. 1976.Google Scholar
- B.H. Thompson and F. Thompson, The New World of Computing. Computer Science Department, California Institute of Technology. 1987.Google Scholar