A database-based rather than a language model-based natural language processing method
Keywords: sentence generation, sentence understanding, relative spatial relationship, Tree-graph hybrid model.
TL;DR: A database-based rather than a language representation model-based natural language generation and understanding method
Abstract: Language models applied to NLP tasks take natural language as the direct modeling object. But we believe that natural language is essentially a way of encoding information, therefore, the object of study for natural language should be the information encoded in language, and the organizational and compositional structure of the information described in language. Based on this understanding, we propose a database-based natural language processing method that changes the modeling object from natural language to the information encoded in natural language. On this basis, the sentence generation task is transformed into read operations implemented on the database and some sentence encoding rules to be followed; The sentence understanding task is transformed into sentence decoding rules and a series of Boolean operations implemented on the database. Our method is closer to the information processing mechanism of the human brain and has excellent interpretability and scalability.
Submission Number: 1249
Loading