On the Main Technical Problems in Construction of a Russian Knowledge Database
FU Xing-shang · Jiefangjun Waiyu Xueyuan xuebao · 2002
Construction of a language knowledge database is critical for enhancement of the efficiency of an NLP system. Word as the basic language unit covers both static categorical knowledge of a language and its dynamic rule knowledge. The Russian language knowledge database under construction takes words as its basic operation units and description objects, and contains word-relevant information represented and stored according to some principles. The main technical problems in the process of construction involve: (1) selection of head words; (2) extraction of semantic information of head words; (3) processing of polysemy; (4) processing of idioms; (5) assignment of Chinese equivalents to Russian head words.