Driving Forces of Natural Language Processing and Logical Semantics at the Age of Artificial Intelligence
Big data-driven natural language processing NLP uses deep learning methods to make the performance of natural language processing increase rapidly.But it does not process on the basis of understanding,so it is not very fit to deal with complex semantics containing human knowledge.The difficulty in overcoming artificial intelligence lies in making the machine truly understand the language.How computers represent,analyze,and understand the semantics of natural language is a key issue for artificial intelligence.The rule-based rationalist orientation of logical semantics can characterize the infinite mechanism of language construction,but without a large amount of real text for natural language,it cannot meet the processing of artificial intelligence.Thus the mainstream NLP is currently reluctant to focus on logical semantics research that is relatively disconnected from the actual situation of natural language.The empirical NLP corpus cannot explain the infinite mechanism of language construction,but it can meet the practical needs of artificial intelligence.Logical semantics should absorb the strengths of empiricism and realize the integration and complementation of rationalist methods and empiricist ones.
Natural Language ProcessingBig DataRich KnowledgeLogical RationalismEmpiricism
万方数据
维普
