YAML Metadata Warning:The task_categories "conversational" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

EcommerceAI Dataset — English E-commerce Customer Service

Dataset Description

A high-quality synthetic English-language dataset of e-commerce customer service conversations with full intent and sentiment annotations.

⚠️ Transparency Notice: This is a synthetic dataset. All conversations are programmatically generated. No real customer data is included. Every record is labeled "data_type": "SYNTHETIC" in metadata.

Dataset Stats

Metric	Value
Conversations	5,000
Dialogue Turns	47,028
Avg Turns/Conversation	9.4
Language	English
Issue Categories	5
Product Categories	10
Intent Classes	12

Issue Categories

Category	Count
Late Delivery	1,000
Wrong Item Received	1,000
Refund Request	1,000
Damaged Item	1,000
Product Inquiry	1,000

Data Schema

{
  "id": "uuid-v4",
  "metadata": {
    "domain": "ecommerce_customer_service",
    "issue_type": "late_delivery",
    "product_category": "electronics",
    "language": "en",
    "data_type": "SYNTHETIC",
    "quality_tier": "enterprise",
    "turns_count": 10
  },
  "conversation": [
    {
      "role": "user",
      "content": "My order hasn't arrived in 7 days.",
      "intent": "late_delivery",
      "sentiment": "negative"
    },
    {
      "role": "agent",
      "content": "I'm sorry to hear that. Let me check your order.",
      "intent": "acknowledge",
      "sentiment": "positive"
    }
  ]
}

Quick Load

# Option 1 — HuggingFace Datasets
from datasets import load_dataset
ds = load_dataset("YOUR_USERNAME/ecommerce-cs-dataset")

# Option 2 — Pure Python
import json
conversations = []
with open('ecommerce_cs_en_synthetic.jsonl') as f:
    for line in f:
        conversations.append(json.loads(line))

# Filter by issue type
refunds = [c for c in conversations 
           if c['metadata']['issue_type'] == 'refund_request']

Use Cases

✅ Fine-tuning LLMs for customer service chatbots
✅ Training intent classifiers (12 classes)
✅ Training sentiment analysis models
✅ Dialogue state tracking research
✅ Augmenting real-world datasets
✅ Testing chatbot pipelines

🔒 Full Dataset (Commercial License)

This repository contains a free sample of 500 conversations.

The full dataset (5,000 conversations, 47K+ turns) with commercial license is available here:

👉 [Get Full Dataset on Sellix → YOUR_SELLIX_LINK]

Includes:

Full 5,000 conversation JSONL
Complete JSON array format
Data card & documentation
Commercial use license

License

Sample (this repo): CC BY-NC 4.0 — free for research & personal use.
Full dataset: Commercial license — see Sellix listing for terms.