A Robust Finite-state Parser for French

This paper describes a robust nite-state parser implemented for French. The parser attaches morpho-syntactic tags to each word and determines clause boundaries. It is a reductionist parser based on nite-state networks and their intersection. We describe essential elements of the rule writing system, and show how it is actually applied to solve various phenomena, such as argument uniqueness, agreement or apposition. We show some results which indicate that the parser can parse technical manuals with high accuracy (in a test sample 95 % of part-of-speech and functional tags were correct). The average number of parses per sentence is very low, more than 92 % of sentences produce less than 4 parses, including the correct one. A test on very long sentences from newspaper corpora and a discussion of errors provide more insight into the parser.