1300 41 7136 9 39794 document 7136 01-Prompt Setting 1-a_RAAC_DefinitionExtraction.py text/plain bc04a4b188c5247346ba1b139a72bfe4 MD5 10717 2025-05-25 19:06:03 http://edata.bham.ac.uk/1300/1/01-Prompt%20Setting%201-a_RAAC_DefinitionExtraction.py 1300 1 1 text/plain text public cc_by_nc_4

01-Prompt Setting 1-a_RAAC_DefinitionExtraction.py

data 7137 9 39795 document 7137 01-Prompt Setting 1-b_RAAC_DefinitionPresenceCheck_BinaryFlag.py text/plain bd37bc332d0c7d0f80ee610bfcc872f4 MD5 7060 2025-05-25 19:06:03 http://edata.bham.ac.uk/1300/2/01-Prompt%20Setting%201-b_RAAC_DefinitionPresenceCheck_BinaryFlag.py 1300 2 2 text/plain text public cc_by_nc_4

01-Prompt Setting 1-b_RAAC_DefinitionPresenceCheck_BinaryFlag.py

data 7138 9 39796 document 7138 02_Prompt Setting 2-b_RAAC_ExplicitMentionCheck_BinaryFlag.py text/x-python 4e62a8f2adfac9c410e18b99860a9518 MD5 2156 2025-05-25 19:06:03 http://edata.bham.ac.uk/1300/2/02_Prompt%20Setting%202-b_RAAC_ExplicitMentionCheck_BinaryFlag.py 1300 2 2 text/x-python text public cc_by_nc_4

02_Prompt Setting 2-b_RAAC_ExplicitMentionCheck_BinaryFlag.py

data 7139 9 39798 document 7139 03_RAAC_DefectExtraction_SevenQuestions.py text/plain c8bf56ac96967e2d3cc05ddb1e5d9374 MD5 11343 2025-05-25 19:06:04 http://edata.bham.ac.uk/1300/3/03_RAAC_DefectExtraction_SevenQuestions.py 1300 3 3 text/plain text public cc_by_nc_4

03_RAAC_DefectExtraction_SevenQuestions.py

data 7142 8 39804 document 7142 04_RAAC_CombineExtractedDefectData.py text/x-python-script 0 2025-05-25 19:06:05 http://edata.bham.ac.uk/1300/5/04_RAAC_CombineExtractedDefectData.py 1300 5 5 text/x-python-script other public cc_by_nc_4

04_RAAC_CombineExtractedDefectData.py

data 7143 8 39806 document 7143 02_Prompt Setting 2-a_RAAC_ExplicitMentionCheck.py text/x-python-script 0 2025-05-25 19:06:05 http://edata.bham.ac.uk/1300/6/02_Prompt%20Setting%202-a_RAAC_ExplicitMentionCheck.py 1300 6 6 text/x-python-script other public cc_by_nc_4

02_Prompt Setting 2-a_RAAC_ExplicitMentionCheck.py

data 7147 5 39831 document 7147 index.html text/html 548bf5acad8587740d925b38a27a1a81 MD5 13997 2025-05-25 20:37:16 http://edata.bham.ac.uk/1300/7/index.html 1300 7 7 text/html text public cc_by_nc_4

index.html

data 7148 5 39833 document 7148 README.md text/plain 8d68874b96392f938289187075bda7bd MD5 7552 2025-05-25 20:37:18 http://edata.bham.ac.uk/1300/8/README.md 1300 8 8 text/plain text public cc_by_nc_4

README.md

readme 7149 5 39842 document 7149 requirements.txt text/plain fe8f6e44b842a52d00c95402f1f822d0 MD5 33 2025-05-25 20:54:40 http://edata.bham.ac.uk/1300/9/requirements.txt 1300 9 9 text/plain text public cc_by_nc_4

requirements.txt

data archive 654 disk0/00/00/13/00 2025-06-02 15:42:19 2026-01-07 08:58:19 2026-01-07 08:58:19 data_collection show duplicate_title_1292 duplicate_title Duplicate title to <a href="http://edata.bham.ac.uk/view/creators/Liu=3ALixu=3A=3A.html"><span class="person_name">Liu, Lixu</span></a> (2025). <a href="http://edata.bham.ac.uk/cgi/users/home?screen=EPrint::View&amp;eprintid=1292"><xhtml:em xmlns:xhtml="http://www.w3.org/1999/xhtml">Research software supporting the publication “Leveraging Large Language Models to Classify and Inspect Defects in Reinforced Autoclaved Aerated Concrete (RAAC)”.</xhtml:em></a> University of Birmingham. DataSet. <a href="https://doi.org/10.25500/edata.bham.00001292" target="_blank">https://doi.org/10.25500/edata.bham.00001292</a>. 2025-06-03 01:30:06 discovered 1 Liu Lixu l.liu.9@bham.ac.uk 0009-0008-8011-496X 10col_ephy Large Language Models (LLMs); Reinforced Autoclaved Aerated Concrete (RAAC); Literature Analysis; Defect Classification; Inspection Methods; Natural Language Processing; High-Throughput Screening The complete codebase is version-controlled and publicly available on GitHub at https://github.com/lixuliu/raac-llm-analysis-code-package. Access to the UBIRA deposit will be made public once the associated paper is published. In the meantime, a project landing page with an overview and usage instructions is available at https://lixuliu.github.io/raac-llm-analysis-code-package/. This site allows readers to explore the code structure, documentation, and example outputs prior to the embargo lift. This deposit contains the full Python code for the LLM-based approach described in “Leveraging Large Language Models to Classify and Inspect Defects in Reinforced Autoclaved Aerated Concrete (RAAC).” It includes six scripts: two parallel prompt modules for RAAC mention detection and definition extraction; a seven-question defect-extraction script; and a data-aggregation script that produces a unified defect database. All scripts are versioned for reproducibility and require Python 3.11+, the Anthropic Claude 3 Opus API, and standard data-analysis libraries. A comprehensive README.md is included, detailing environment setup, dependency installation, API key configuration, and step-by-step execution instructions. The code is mirrored on GitHub for ongoing collaboration and version tracking. An interactive project overview and navigation interface is also provided via index.html on the project’s GitHub Pages site. 2025-06-02 published University of Birmingham https://doi.org/10.25500/edata.bham.00001300 Software research-data@contacts.bham.ac.uk School of Engineering, Department of Civil Engineering other Research England Imfuna ltd C-DICE Industrial Secondement en yes yes yes TRUE Primary code maintained in a public GitHub repository for versioning and collaboration https://github.com/lixuliu/raac-llm-analysis-code-package.git Liu Lixu l.liu.9@bham.ac.uk; lixu@verdemetrix.com web https://github.com/lixuliu/raac-llm-analysis-code-package.git IsVersionOf web https://lixuliu.github.io/raac-llm-analysis-code-package IsDescribedBy Ran two parallel Claude 3 Opus prompts (“definition extraction” and “explicit mention check”) over the provided text corpus. Applied the seven-question defect-extraction prompt to the validated subset. Merged all resulting CSV files into one consolidated defect database via the aggregation script. en