Device and method for automatically generating structural rules of building codes

Abstract

The invention discloses a device for automatically generating structured building codes based on semantic analysis, which includes a coding module for building code domain features, a generating module with SNL syntax constraints, and a two-stage generating module, and is characterized in that: the building code domain features The coding module obtains the clause semantic vector according to the natural language description; the generation module with SNL grammatical constraints realizes the generation function with grammatical constraints; The generation process of the sentence pattern template is transformed into the task of selecting the sentence template and the task of filling the details of the sentence pattern, so as to realize the restriction on the generation process of the decoder; the two-stage generation module provides the function of converting natural language rules into SNL; the two-stage generation module will The encoding method is fused with the generative method with grammatical constraints to obtain a two-stage generative model to reduce the errors caused by overfitting of a single model.

Description

technical field [0001] The invention relates to a method for automatic generation of structured building codes, in particular to a device and method for realizing from natural language to structured building codes using semantic parsing technology, which is applied to the structured writing stage of building codes and building information models. Review, design stage. Background technique [0002] As the level of digitization in the field of architectural design continues to deepen, the automatic review technology of architectural drawings is becoming more and more important. In automatic review technology, structured specification is the basic data support. This is because the rules described in natural language are not enforceable. At this stage, the structuring of current building codes is mainly done manually. This method mainly has the problems of long development cycle, high personnel requirements and high collaboration cost. Therefore, tools for automatic structur...

AI Technical Summary

Problems solved by technology

[0009] (1) Among the existing semantic analysis tasks, SQL and code generation are more common, and the methods include sketch filling and reinforcement learning methods, but they are mainly designed for SQL / Python, and the grammatical structure of these languages ​​​​is quite different from SNL

[0010] (2) The existing semantic analysis technology relies less on the background knowledge of a specific field, and there are a large number of proper nouns in the construction field, which need to be distinguished, otherwise the generated results will be poor

[0011] (3) Existing constrained generation techniques are often designed for deep recursive structures, which do not match the actual situation of SNL

If you forcibly support a deeper structure, there will be a problem of under-learning

Images