Predictive Control Parser: Extended LINGOL

Hozumi Tanaka, Taisuke Sato, Fumio Motoyoshi

Abstract: We have developed a flexible parser called Extended LINGOL which is an extended version of Pratt's LINGOL. Although original LINGOL has several advantages, the results of our experiments revealed that it has some deficiencies. One of them is that each augmented context-free rule does not enable us to examine their applicability in case of some situations. Another is that the morpheme analyzer of original LINGOL is too simple to analyze aggulutinative languages such as Japanese. With respect to the former problem, we have introduced predictive control mechanism which has offered us flexible methods to implement a precise grammar with a small set of rules. Furthermore, as the mechanism avoids generating unnecessary partially parsed trees we have obtained high speed parsing. On the other hand, with respect to the latter problem, we have implemented a more powerful morpheme analyzer.