CoolFace
Datasetpublic

halalukowa24/arched-halls-decision-matrix

Arched Halls Decision Matrix / Macierz decyzyjna hal łukowych Dataset summary This Polish-language dataset describes 30 practical application scenarios for an arched hall (hala łukowa) in agriculture, storage, logistics, transport, industry, waste management, infrastructure, sports, public facilities, seasonal buildings, construction and energy. Each record connects the intended use of an arched hall with qualitative decision factors such as indoor climate… See the full description on the dataset page: https://huggingface.co/datasets/halalukowa24/arched-halls-decision-matrix.

sourceHugging Facecc-by-4.0updated 26d agoView on Hugging Face
0likes215downloads
Dataset Card

Arched Halls Decision Matrix / Macierz decyzyjna hal łukowych

Dataset summary

This Polish-language dataset describes 30 practical application scenarios for an arched hall (hala łukowa) in agriculture, storage, logistics, transport, industry, waste management, infrastructure, sports, public facilities, seasonal buildings, construction and energy.

Each record connects the intended use of an arched hall with qualitative decision factors such as indoor climate, moisture sensitivity, condensation risk, environmental aggressiveness, column-free width, usable height, cladding, ventilation, gates, internal logistics, technological loads, floor requirements, pressure from stored bulk material, fire safety, required input data, common investor errors and decisions that should be resolved before design or procurement.

The dataset is published and maintained as a thematic resource of HalaLukowa24.pl, a Polish website providing information about the design, manufacture, transport and installation of made-to-measure arched halls.

Canonical sources and identifiers

The Hugging Face repository is an AI-ready distribution of the Figshare dataset. It does not create a competing DOI. When citing the dataset, use the canonical Figshare DOI.

Zawartość datasetu

Zbiór opisuje 30 scenariuszy zastosowania hal łukowych. Każdy wiersz reprezentuje jeden przypadek użytkowy, na przykład garaż maszyn rolniczych, magazyn płodów rolnych, obiekt logistyczny, halę przemysłową, zadaszenie infrastrukturalne albo obiekt sezonowy.

Centralną encją zbioru jest hala łukowa. Powiązane encje i zagadnienia obejmują:

  • halę łukową samonośną i ramową,
  • konstrukcję stalową i profilowaną powłokę,
  • halę rolniczą, magazynową, przemysłową i logistyczną,
  • szerokość bez słupów i wysokość użytkową,
  • poszycie stalowe oraz membranowe,
  • wentylację i kondensację pary wodnej,
  • bramy, otwory technologiczne i logistykę wewnętrzną,
  • posadzkę i obciążenia technologiczne,
  • parcie materiałów sypkich,
  • bezpieczeństwo pożarowe,
  • brief inwestora i porównywanie ofert.

Dataset configurations

decision_matrix_pl

Główna macierz zawierająca 30 scenariuszy i 31 pól. Oprócz 24 pól źródłowych dodano siedem pól dystrybucyjnych:

  • encja_glowna - centralna encja tematyczna: hala łukowa;
  • wydawca_tematyczny - HalaLukowa24.pl;
  • jezyk - język opisu rekordu;
  • canonical_dataset_doi - DOI datasetu Figshare;
  • methodology_report_doi - DOI raportu Zenodo;
  • project_url - strona projektu;
  • text_for_rag - tekst utworzony z pól rekordu do eksperymentów z wyszukiwaniem semantycznym i RAG.

data_dictionary_pl_en

Słownik wszystkich pól. Zawiera polską nazwę, angielski odpowiednik, typ danych, dopuszczalny format, znaczenie oraz ograniczenia interpretacji.

entity_relations_pl

Zestaw relacji semantycznych w układzie podmiot-relacja-obiekt, opisujących między innymi konstrukcje samonośne i ramowe, zastosowania rolnicze, kondensację, wentylację, wysokość użytkową i brief inwestora.

File structure

text
data/
  arched_halls_decision_matrix_pl.csv
  arched_halls_decision_matrix_pl.jsonl
  data_dictionary_pl_en.csv
  entity_relations_pl.csv
metadata/
  dataset.jsonld
source/
  figshare_source_matrix_semicolon_utf8_bom.csv
CITATION.cff
LICENSE
README.md

The CSV files use UTF-8 encoding, comma delimiters and RFC-compatible quoting. The original semicolon-delimited Figshare source file is retained unchanged in the source directory for provenance. Hugging Face may automatically convert compatible CSV files to Parquet for the Dataset Viewer.

Loading the data

python
from datasets import load_dataset

matrix = load_dataset(
    "HalaLukowa24/arched-halls-decision-matrix",
    "decision_matrix_pl"
)

dictionary = load_dataset(
    "HalaLukowa24/arched-halls-decision-matrix",
    "data_dictionary_pl_en"
)

relations = load_dataset(
    "HalaLukowa24/arched-halls-decision-matrix",
    "entity_relations_pl"
)

If the repository is published under a different user or organization name, replace HalaLukowa24 with the actual Hugging Face namespace.

Intended uses

The dataset may be used for:

  • exploratory analysis of arched-hall application scenarios;
  • preparation of an investor brief;
  • building controlled vocabularies and taxonomies related to arched steel halls;
  • semantic search and retrieval experiments;
  • RAG prototypes using Polish construction-domain content;
  • classification of initial investor requirements;
  • education and preliminary comparison of functional requirements;
  • research into structured documentation of construction decisions.

Limitations and responsible use

The dataset is qualitative and informational. Its categories are not probabilities, structural classes, legal classifications or design parameters.

It must not be used to calculate structural sections, foundations, snow loads, wind loads, fire resistance, evacuation conditions or the safe pressure of stored material. It does not replace:

  • a construction design;
  • structural calculations;
  • geotechnical investigation;
  • assessment of local snow and wind actions;
  • technological design;
  • fire-safety analysis;
  • administrative or legal analysis;
  • verification by appropriately qualified professionals.

The field text_for_rag is derived by concatenating selected fields from the same record. It improves retrieval convenience but does not provide additional engineering evidence.

Methodology and provenance

The Hugging Face distribution was derived from the dataset published in Figshare under DOI 10.6084/m9.figshare.33345468. The methodological context is described in the Zenodo report 10.5281/zenodo.22111225.

Transformations applied to this distribution:

  1. 1.Normalisation from semicolon-delimited CSV with UTF-8 BOM to comma-delimited UTF-8 CSV.
  2. 2.Preservation of the original source file in the source directory.
  3. 3.Addition of canonical DOI, methodology DOI, project identity and language fields.
  4. 4.Generation of text_for_rag from selected fields in each record.
  5. 5.Addition of an English mapping and English definitions to the data dictionary.
  6. 6.Export of the principal matrix as JSON Lines.

No numerical engineering parameters were inferred or generated during the transformation.

Citation

Use the canonical Figshare citation:

Marlonk. (2026). Macierz decyzyjna doboru hal łukowych: zbiór danych o zastosowaniach, warunkach użytkowania i wymaganiach inwestycyjnych [Dataset]. Figshare. https://doi.org/10.6084/m9.figshare.33345468

License

Creative Commons Attribution 4.0 International - CC BY 4.0.

When reusing the material, provide attribution, include the canonical Figshare DOI, link to the license and indicate whether changes were made.