作者: Franz Geiselbrechtinger , Tahar Kechadi , John A. Fitzgerald
DOI:
关键词: Parsing 、 Symbol (chemistry) 、 Syntax (programming languages) 、 Handwriting recognition 、 Natural language processing 、 Basis (linear algebra) 、 Artificial intelligence 、 Computer science 、 Process (engineering) 、 Fuzzy logic 、 Binary expression tree
摘要: Including mathematical expressions in documents can be a tiresome and difficult process. A recognition system for handwritten would greatly simplify the task, enabling users to enter an intuitive manner. In this paper powerful yet efficient approach expression is presented. Our built upon traditional shift-reduce parsing methods which have linear complexity, thus providing necessary syntax checking also basis system. Fuzzy logic introduced basic framework, appropriate given ambiguous nature of symbol identities spatial relationships expressions. Multiple parses are explored whenever strong ambiguities arise, most likely tree selected as result.