toeic-vocab-tw / README.md
LEE-WHITE's picture
Duplicate from kknono668/toeic-vocab-tw
8850458 verified
metadata
pretty_name: 完整 TOEIC 單字庫 (English–Traditional Chinese)
license: cc-by-4.0
language:
  - en
  - zh
language_bcp47:
  - zh-Hant
configs:
  - config_name: default
    data_files:
      - data/toeic_vocabulary_view.json
tags:
  - toeic
  - vocabulary
  - bilingual
  - english-chinese
  - traditional-chinese
  - business
  - education
task_categories:
  - other
size_categories:
  - 1K<n<10K

完整 TOEIC 單字庫(English–Traditional Chinese)

一份面向 TOEIC 情境的雙語單字庫,涵蓋商務、辦公、會議、客戶服務、財務會計、行銷銷售、採購物流、法律合規、資訊技術等多個常見領域。資料以 JSON 提供,並依重要度劃分 5★–1★ 等級,附詞性、詞形變化與中英雙語例句,便於教學、備考、應用開發與語言學習研究。

Dataset Summary

  • Title: 完整 TOEIC 單字庫
  • Version: 110.5
  • Total Words: 9,537
  • Language Pair: English ↔ Traditional Chinese (繁體中文)
  • Structure: 單一 JSON 陣列(每列一個詞條)+ data/metadata.json

Files

  • data/toeic_vocabulary.json:主資料檔(頂層為陣列,每列一個詞條)
  • data/metadata.json:資料集中繼資料(title/version/last_updated/total_words…)

可選(仍為 JSON,便於 Viewer 預覽):

  • data/toeic_vocabulary_view.json:Viewer 友善版(巢狀欄位以字串存放,如 word_forms_jsonexamples_json,方便 HF 轉成表格)

Schema

  • 主檔(data/toeic_vocabulary.json)為「頂層陣列」,每個詞條包含:

  • english_word:英文單字

  • chinese_definition:中文定義

  • star_rating:整數(1–5)

  • category:所屬主題分類

  • parts_of_speech:詞性陣列(如 noun/verb/adj/adv 等)

  • word_forms:依詞性細分的詞形(例:名詞複數、動詞三單/過去式/分詞/動名詞等)

  • examples:例句陣列(englishchinese

  • importance_group:原始分組標籤(如 5_stars_essential

  • 中繼資料(data/metadata.json):

    • titleversiondescriptiontotal_wordscreation_datelanguage_pair
    • categoriesstar_systemlast_updated

Star System 說明

  • 5★ 核心商業詞彙:最高優先級,TOEIC 必備
  • 4★ 重要專業詞彙:商務場合常用
  • 3★ 職場/學術詞彙:有助提升語言能力
  • 2★ 有用支援詞彙:擴展詞彙量
  • 1★ 一般詞彙:建議了解

Example Instance

{
  "english_word": "answer",
  "chinese_definition": "回覆;答覆;回應",
  "star_rating": 5,
  "category": "溝通互動",
  "parts_of_speech": ["noun", "verb"],
  "word_forms": {
    "noun": {"plural": "answers"},
    "verb": {
      "third_person_sg": "answers",
      "past": "answered",
      "past_participle": "answered",
      "present_participle": "answering"
    }
  },
  "examples": [
    {"english": "Could you please answer the questionnaire before the meeting?", "chinese": "請在會議前填寫問卷並回覆。"},
    {"english": "The applicant needs to answer all the questions truthfully.", "chinese": "申請人需要誠實地回答所有問題。"}
  ],
  "importance_group": "5_stars_essential"
}

Usage

Local (JSON)

import json
from pathlib import Path

path = Path("data/toeic_vocabulary.json")
rows = json.loads(path.read_text(encoding="utf-8"))  # 頂層為陣列

print(len(rows), "words total")
# 篩選 5★ 詞條
top_words = [r for r in rows if r.get("importance_group") == "5_stars_essential"]
print(len(top_words), "top words")
print(top_words[0]["english_word"], top_words[0]["chinese_definition"])

After Upload (datasets)

上傳到 Hugging Face 後,可使用 datasets 套件讀取:

from datasets import load_dataset

# 直接載入主檔(頂層陣列 JSON;保留巢狀欄位)
ds = load_dataset(
    "json",
    data_files="https://huggingface.co/datasets/kknono668/toeic-vocab-tw/resolve/main/data/toeic_vocabulary.json",
    split="train",
)

# 若要簡化成純表格欄位,可改載入 viewer 友善版(巢狀欄位已字串化)
# ds = load_dataset(
#   "json",
#   data_files="https://huggingface.co/datasets/kknono668/toeic-vocab-tw/resolve/main/data/toeic_vocabulary_view.json",
#   split="train",
# )

提示:若要轉為扁平表格(如 Pandas DataFrame),可將各重要度群組展開並加上 importance_group 欄位。

Intended Uses

  • TOEIC 應試準備、語言教學與練習
  • 商務英語學習應用(查詢、測驗、卡片、練習題生成)
  • 詞彙表與例句檢索工具

Limitations

  • 定義與例句聚焦於職場/商務情境,對一般語境未必全面
  • 詞性與詞形為常見用法彙整,未涵蓋所有罕見變化

License

本資料集以 CC BY 4.0 釋出。請在使用時適當標註來源與作者,並附上授權連結。

Citation

如果你在研究或產品中使用了本資料集,請引用:

@misc{toeic_vocab_2025,
  title  = {完整 TOEIC 單字庫 (English–Traditional Chinese)},
  year   = {2025},
  note   = {CC BY 4.0},
  howpublished = {Hugging Face Dataset}
}

Changelog

  • v34.8: 初版公開;9,855 詞條;依 5★–1★ 分組;含詞性、詞形、例句與分類。