Datasets:
The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: TypeError
Message: Couldn't cast array of type
struct<太陽光パネルメーカー別品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 蓄電池メーカー別品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, スイッチ・コンセント品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, カーテン・ブラインド全種類: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 舗装材詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 仮設・建設機械リース詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 工事保険・保証費用: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 残土・産廃品目別処分費: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, コンクリート添加剤・特殊材: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 造作家具・オーダー家具: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 植栽・造園資材詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 住宅性能評価・申請費用: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 特殊塗装・防錆・耐熱塗装: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>>
to
{'仮設工事': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '基礎工事': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('float64'), 'memo': Value('string')}), '木構造材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '構造用合板・面材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '金物・ビス・釘': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '断熱材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '屋根材・防水': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '外壁材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '木製建具・サッシ': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '内装下地材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '内装仕上げ材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '水回り設備': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '給排水配管材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '電気配線材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '外構・仕上げ': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '塗装・コーキング材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '資材高騰情報_2026年4月': {'総合指数_2021年比': Value('string'), '直近_2024年12月比': Value('string'), '特に高騰している品目': List(Value('string')), '品薄・納期遅延品目': List(Value('string')), '出典': Value('string')}}
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
return get_rows(
^^^^^^^^^
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/utils.py", line 77, in get_rows
rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
for key, pa_table in self._iter_arrow():
^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 299, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2255, in cast_table_to_schema
cast_array_to_feature(
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 1804, in wrapper
return pa.chunked_array([func(chunk, *args, **kwargs) for chunk in array.chunks])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2101, in cast_array_to_feature
raise TypeError(f"Couldn't cast array of type\n{_short_str(array.type)}\nto\n{_short_str(feature)}")
TypeError: Couldn't cast array of type
struct<太陽光パネルメーカー別品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 蓄電池メーカー別品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, スイッチ・コンセント品番: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, カーテン・ブラインド全種類: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 舗装材詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 仮設・建設機械リース詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 工事保険・保証費用: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 残土・産廃品目別処分費: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, コンクリート添加剤・特殊材: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 造作家具・オーダー家具: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 植栽・造園資材詳細: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 住宅性能評価・申請費用: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>, 特殊塗装・防錆・耐熱塗装: list<item: struct<name: string, unit: string, unit_price: int64, memo: string>>>
to
{'仮設工事': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '基礎工事': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('float64'), 'memo': Value('string')}), '木構造材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '構造用合板・面材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '金物・ビス・釘': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '断熱材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '屋根材・防水': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '外壁材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '木製建具・サッシ': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '内装下地材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '内装仕上げ材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '水回り設備': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '給排水配管材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '電気配線材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '外構・仕上げ': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '塗装・コーキング材': List({'name': Value('string'), 'unit': Value('string'), 'unit_price': Value('int64'), 'standard_qty_30tsubo': Value('int64'), 'memo': Value('string')}), '資材高騰情報_2026年4月': {'総合指数_2021年比': Value('string'), '直近_2024年12月比': Value('string'), '特に高騰している品目': List(Value('string')), '品薄・納期遅延品目': List(Value('string')), '出典': Value('string')}}Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
JCCDB: 日本建設費オープンデータベース
日本の建設・リフォーム工事の参考単価データベース。木造2階建て30坪の標準仕様について、工事カテゴリ別に参考単価と標準数量をまとめたもの。バージョン 2026.04。
要点
JCCDBは、日本の建設工事の項目を、カテゴリ別に整理し、各項目に参考単価と標準数量を付けたオープンなデータベースです。この公開版は、木造2階建て30坪の標準仕様を対象としています。
施主が「この見積もりは妥当か」を自分で確かめられるように、工事を項目単位に分解し、一行ずつ突き合わせられる材料を提供します。
このデータセットは何か
日本の建設・リフォーム見積もりは、業者ごとにバラバラの書き方で作られます。同じ工事が、ある見積もりでは「一式」1行に、別の見積もりでは20行の内訳に化けます。これでは施主が公正な比較をするのはほぼ不可能で、適正な見積もりと水増しされた見積もりを見分けられない最大の原因のひとつになっています。
JCCDBは、その欠けていた共通の物差しを提供します。工事項目をカテゴリ別に整理し、各項目に参考単価と、30坪標準仕様での標準数量を付けてあります。これがあれば、手元の見積もりを項目単位に分解し、参考値と一行ずつ突き合わせて、極端に外れた行に気づくことができます。
単価について
このデータセットには単価が入っています。これは意図的なものです。
各項目の unit_price は、2026年4月時点の参考単価です。資材の市況(資材高騰など)を反映した、木造2階建て30坪の標準仕様を前提とする目安です。standard_qty_30tsubo は、その標準仕様で必要になるおおよその数量です。
これらは比較の出発点であって、確定見積もりではありません。実際の価格は、地域、時期、資材相場、業者の諸経費、工事の正確な範囲によって変わります。参考単価と数量を使って自分の見積もりを突き合わせ、外れに気づくこと。それがこのデータセットの使い方です。
データ構造
データ本体は13分割のJSONファイル(zaisai-db.json から zaisai-db13.json)です。
各ファイルの最上位は次の3つを持ちます。
- version: データのバージョン(例 "2026.04")
- note: そのバージョンの前提や注記
- categories: 工事カテゴリ名をキーとするオブジェクト
categories の各カテゴリ(例 仮設工事)は、工事項目オブジェクトの配列です。各オブジェクトは次のキーを持ちます。
- name: 項目名(例 養生テープ (50mm×25m))
- unit: 数量単位(例 巻、枚、m2)
- unit_price: 参考単価(円、2026.04時点)
- standard_qty_30tsubo: 木造2階建て30坪標準仕様での標準数量
- memo: 補足(調達状況や使用場面などの注記)
来歴と方法
JCCDBは、30年以上の建設現場での実務経験から編纂したものです。編纂者は15歳で大工として現場に入り、現場監督・施工管理を経て、独学でソフトウェアとAIの開発を行いながら本データベースを構築しました。
バージョンには、各バージョンがいつ存在し、後から改変されていないことを誰でも確認できるよう、公開かつ検証可能なタイムスタンプがあります。
- SSRN abstract 6738701
- engrXiv DOI 10.31224/7007
- Zenodo DOI 10.5281/zenodo.20019573
- ビットコイン・ブロックチェーンによるタイムスタンプ(ブロック949356)
想定される用途
- 手元の見積もりを参考単価と突き合わせ、妥当性を確かめる。
- 建設経済学、数量積算、信用財(credence goods)市場の研究。
- 共通の項目構造と参考単価の上に、中立なツール(積算アシスタント、比較ツール、AIアプリ)を作る。
- 買い手と業者の情報の非対称を減らす消費者保護の取り組み。
制約と適用範囲
- 単価は2026年4月時点の参考値であり、確定見積もりではありません。地域・時期・範囲で変動します。
- 木造2階建て30坪の標準仕様を前提としています。
- 日本の建設実務に固有のものです。
- 専門家による見積もりや現地調査の代わりにはなりません。
ライセンス
クリエイティブ・コモンズ 表示 4.0 国際(CC-BY 4.0)で公開。適切なクレジットを表示すれば、商用を含め共有・改変が可能です。
引用
@misc{oga_jccdb_2026,
author = {Oga, Toshikatsu},
title = {JCCDB: Japan Construction Cost open DataBase},
year = {2026},
publisher = {Zenodo},
doi = {10.5281/zenodo.20019573},
url = {https://doi.org/10.5281/zenodo.20019573}
}
著者
大賀俊勝(Toshikatsu Oga)。The HORIZ音s株式会社 代表取締役。
- ORCID: 0009-0000-9180-903X
- Web of Science ResearcherID: QAY-0724-2026
- プロジェクトサイト: https://shield.the-horizons-innovation.com
研究プログラム
JCCDBは、建設における情報の非対称を扱う、より大きな一連の研究の基盤です。公開済みの関連成果は次のとおりです。
JCCDB v1.2 基盤論文。再現可能なLLMベースの建設費診断のための、暗号学的監査ハッシュとマクロ経済的な価格補正を扱います。SSRN abstract 6738701, engrXiv DOI 10.31224/7007, Zenodo DOI 10.5281/zenodo.20195706。
VRQ(検証付きリフォーム見積もり)枠組み。住宅リフォームにおける価格分散と信用財を扱い、第三者検証による見積もり認証を提案します。SSRN abstract 6807738, engrXiv DOI 10.31224/7126, Zenodo DOI 10.5281/zenodo.20372953。
非標準工事へのJCCDB拡張。標準的な価格参照が無い工事で起きる「幻の過剰請求」を定式化し、メーカーや施工業者の一次資料を参照層に組み込む補正法を提案します(参照層を170から176カテゴリへ拡張)。Zenodo DOI 10.5281/zenodo.20370618。
一般向け大規模言語モデルのベンチマーク。40問のリフォーム費用質問で、GPT-4o・Gemini と構造化エンジンを比較し、汎用モデルが再現性を欠く一方で構造化エンジンが決定的であることを示します。SSRN abstract 6872819, 再現データ Zenodo DOI 10.5281/zenodo.20522847。
方法論宣言(PTKA)は、ビットコインの OpenTimestamps で時刻認証され、独立に検証できます。Zenodo DOI 10.5281/zenodo.20175219。
これらは合わせて一本の問いを形づくります。建設費の中立でオープンな物差しを作り、なぜ市場が買い手を裏切るのかを説明する。
(English) JCCDB: Japan Construction Cost open DataBase
An open reference database of Japanese construction cost items. This release provides reference unit prices and standard quantities for a standard 30-tsubo (about 100 square metre) wood-frame two-story house, organised by work category. Version 2026.04.
What this dataset is
Construction and renovation estimates in Japan are written by different contractors in different ways. The same job can appear as one lump-sum line in one estimate and as twenty itemised lines in another. That makes honest comparison almost impossible for a homeowner. JCCDB provides the missing common yardstick: work items grouped by category, each with a reference unit price and a standard quantity for a standard 30-tsubo build, so an estimate can be broken down and checked line by line.
About the prices
This dataset contains prices on purpose. Each unit_price is a reference value as of April 2026, reflecting current material costs, for a standard 30-tsubo wood-frame house. standard_qty_30tsubo is the rough quantity needed for that standard spec. These are a starting point for comparison, not a fixed quote. Real prices vary by region, timing, market, contractor overhead, and exact scope. The point is to compare your own estimate against these reference figures and notice the lines that are far off.
Dataset structure
The data is split across 13 JSON files (zaisai-db.json to zaisai-db13.json). Each file has three top-level keys: version, note, and categories. categories is an object keyed by work-category name (for example temporary works). Each category holds an array of item objects with these keys:
- name: item name
- unit: unit of measurement
- unit_price: reference unit price in JPY, as of 2026.04
- standard_qty_30tsubo: standard quantity for a 30-tsubo wood-frame house
- memo: notes (procurement status, typical use, and so on)
Provenance and methodology
JCCDB was compiled from more than 30 years of hands-on construction site experience. The compiler began as a carpenter at age 15, worked as a site supervisor and construction manager, and later built the database while teaching himself software and AI work. Public, verifiable timestamps: SSRN abstract 6738701, engrXiv DOI 10.31224/7007, Zenodo DOI 10.5281/zenodo.20019573, and a Bitcoin blockchain timestamp anchor (block 949356).
Intended uses
Checking your own estimate against reference unit prices; research in construction economics, quantity surveying, and credence-goods markets; building neutral tooling on a shared structure of items and reference prices; consumer protection work that reduces information asymmetry.
Limitations and scope
Prices are reference values as of April 2026 and are not a quote; they vary by region, timing, and scope. Assumes a standard 30-tsubo wood-frame two-story house. Specific to construction practice in Japan. Not a substitute for a professional estimate or site inspection.
License
Creative Commons Attribution 4.0 International (CC-BY 4.0).
Research program
JCCDB is the foundation for a wider body of work on information asymmetry in construction. The publicly available outputs are:
- JCCDB v1.2 foundation paper, covering a cryptographic audit hash and macroeconomic price correction for reproducible LLM-based construction cost diagnostics. SSRN abstract 6738701, engrXiv DOI 10.31224/7007, Zenodo DOI 10.5281/zenodo.20195706.
- Verified Renovation Quote (VRQ) framework: price dispersion and credence goods in residential renovation, proposing third-party verified quote certification. SSRN abstract 6807738, engrXiv DOI 10.31224/7126, Zenodo DOI 10.5281/zenodo.20372953.
- A primary-source-based extension of JCCDB to non-standardized work, formalizing the phantom-overcharge failure mode and a correction method that incorporates manufacturer and installer primary sources into the reference layer (expanding it from 170 to 176 categories). Zenodo DOI 10.5281/zenodo.20370618.
- A benchmark of consumer large language models: across 40 renovation cost questions, GPT-4o and Gemini are compared against a structured engine, showing the general-purpose models to be non-reproducible while the structured engine is deterministic. SSRN abstract 6872819, replication data Zenodo DOI 10.5281/zenodo.20522847.
The methodology declaration (PTKA) is anchored to Bitcoin via OpenTimestamps and is independently verifiable. Zenodo DOI 10.5281/zenodo.20175219.
Author: Toshikatsu Oga, Representative Director, The HORIZ音s株式会社. ORCID 0009-0000-9180-903X. ResearcherID QAY-0724-2026. https://shield.the-horizons-innovation.com
- Downloads last month
- 133