Generative artificial intelligence (AI) construction specification interface
A method and system provide the ability to process a construction domain query. A natural language user query is obtained within a construction software system. The user query is pre-processed to validate the query. Text from the query is embed into search vectors for a semantic search. A data source having multiple different sections is obtained. The semantic search is performed within each of the sections and identifies semantically relevant sections. The relevant sections are consolidated into a contextual data prompt that is input into an LLM. The LLM, which is trained based on construction data, generates a response that identifies the relevant sections. The response and an identification of the relevant sections is output.
1 . A computer-implemented method for processing a construction domain query, comprising:
obtaining a user query comprising a natural language query within a construction software system;
pre-processing the user query to validate the query;
embedding text from the preprocessed user query into search vectors for a semantic search;
obtaining a data source comprising multiple different sections, wherein the data source comprises a vector database with stored vectors representing the multiple different sections;
performing the semantic search within each of the multiple different sections, based on the search vectors and the stored vectors, wherein the semantic search identifies semantically relevant sections of the multiple different sections;
consolidating the semantically relevant sections into a contextual data prompt;
utilizing a large language model (LLM) to generate a response based on the contextual data prompt, wherein the LLM is trained based on construction data, wherein the construction data comprises multiple different data that is siloed within the construction software system, and wherein the response identifies the semantically relevant sections; and
outputting the response and identification of the semantically relevant sections.
2 . The computer-implemented method of claim 1 , wherein the pre-processing comprises correcting typographical errors, verifying a quality and safety of user input, and rephrasing and complementing the user query.
3 . The computer-implemented method of claim 1 , further comprising:
optimizing information retrieval to improve the semantic search.
4 . The computer-implemented method of claim 3 , wherein the optimizing utilizes an optimization strategy comprising:
a chunking method that chunks the data source into text segments.
5 . The computer-implemented method of claim 3 , wherein the optimizing utilizes an optimization strategy comprising:
a similarity score that measures text relevancy.
6 . The computer-implemented method of claim 3 , wherein the optimizing utilizes an optimization strategy comprising:
a retriever that retrieves relevant text using a searching strategy.
7 . The computer-implemented method of claim 3 , wherein the optimizing utilizes an optimization strategy comprising:
a searching strategy that determines where to search and what chunks to retrieve.
8 . The computer-implemented method of claim 1 , wherein the outputting utilizes a text generation optimization that comprises a safety check of the response.
9 . The computer-implemented method of claim 1 , wherein the outputting utilizes a text generation optimization that comprises a hallucination check of the response.
10 . A computer-implemented system for processing a construction domain query, comprising:
(a) a computer having a memory;
(b) a processor executing on the computer;
(c) the memory storing a set of instructions, wherein the set of instructions, when executed by the processor cause the processor to perform operations comprising:
(1) obtaining a user query comprising a natural language query within a construction software system;
(2) pre-processing the user query to validate the query;
(3) embedding text from the preprocessed user query into search vectors for a semantic search;
(4) obtaining a data source comprising multiple different sections, wherein the data source comprises a vector database with stored vectors representing the multiple different sections;
(5) performing the semantic search within each of the multiple different sections, based on the search vectors and the stored vectors, wherein the semantic search identifies semantically relevant sections of the multiple different sections;
(6) consolidating the semantically relevant sections into a contextual data prompt;
(7) utilizing a large language model (LLM) to generate a response based on the contextual data prompt, wherein the LLM is trained based on construction data, wherein the construction data comprises multiple different data that is siloed within the construction software system, and wherein the response identifies the semantically relevant sections; and
(8) outputting the response and identification of the semantically relevant sections.
11 . The computer-implemented system of claim 10 , wherein the pre-processing comprises correcting typographical errors, verifying a quality and safety of user input, and rephrasing and complementing the user query.
12 . The computer-implemented system of claim 10 , further comprising:
optimizing information retrieval to improve the semantic search.
13 . The computer-implemented system of claim 12 , wherein the optimizing utilizes an optimization strategy comprising:
a chunking method that chunks the data source into text segments.
14 . The computer-implemented system of claim 12 , wherein the optimizing utilizes an optimization strategy comprising:
a similarity score that measures text relevancy.
15 . The computer-implemented system of claim 12 , wherein the optimizing utilizes an optimization strategy comprising:
a retriever that retrieves relevant text using a searching strategy.
16 . The computer-implemented system of claim 12 , wherein the optimizing utilizes an optimization strategy comprising:
a searching strategy that determines where to search and what chunks to retrieve.
17 . The computer-implemented system of claim 10 , wherein the outputting utilizes a text generation optimization that comprises a safety check of the response.
18 . The computer-implemented system of claim 10 , wherein the outputting utilizes a text generation optimization that comprises a hallucination check of the response.