Dataset Viewer
Auto-converted to Parquet Duplicate
dataset_info
dict
conversations
list
{ "domain": "financial support", "total_conversations": 100, "turns_per_conversation": 11, "customer_types": [ "frustrated user", "confused beginner", "impatient executive", "elderly user", "tech-savvy user", "calm and patient user" ], "version": "1.0" }
[ { "id": 1, "domain": "financial support", "problem": "account closure delayed", "customer_type": "frustrated user", "dialogue": [ { "role": "customer", "text": "Hello, I'm concerned about my account. Account closure delayed." }, { "role": "agent", ...

YAML Metadata Warning: empty or missing yaml metadata in repo card

Check out the documentation for more information.

Financial Support Conversations Dataset

This dataset contains 100 realistic financial support conversations between customers and agents. Each dialogue is 11 turns long and covers common banking and financial issues such as suspicious charges, failed wire transfers, fraud disputes, account management, and transaction problems. It is ideal for training AI assistants, chatbots, and customer support models for banks and financial institutions.

Dataset Structure

Each conversation is stored as a JSON object with the following fields:

  • id: Unique identifier for the conversation.
  • domain: Always "financial support" for this dataset.
  • problem: Short description of the customer's issue.
  • customer_type: Customer persona (e.g., "frustrated user", "tech-savvy user").
  • dialogue: List of alternating customer/agent messages. Each message has:
    • role: "customer" or "agent".
    • text: The message content.
  • resolution: One-sentence summary of how the issue was resolved.

Example Conversation

{
  "id": 1,
  "domain": "financial support",
  "problem": "suspicious charges on account",
  "customer_type": "frustrated user",
  "dialogue": [
    {
      "role": "customer",
      "text": "I just checked my account and there are three charges I did not authorize. I'm really worried."
    },
    {
      "role": "agent",
      "text": "I understand this is alarming. Your security is our top priority. Can I verify the last 4 digits of your card?"
    },
    {
      "role": "customer",
      "text": "Last 4 digits are 7734. The charges appeared this morning."
    },
    {
      "role": "agent",
      "text": "Thank you. I've accessed your account securely. I can see three unauthorized transactions. I'm freezing those transactions and blocking the card immediately."
    },
    {
      "role": "customer",
      "text": "Good. Is my money safe? Can I get those charges reversed?"
    },
    {
      "role": "agent",
      "text": "Your funds are protected. I'm filing a fraud dispute for all three charges right now. You'll receive a provisional credit within 24 hours while we investigate."
    },
    {
      "role": "customer",
      "text": "A provisional credit — so I won't be out of pocket while you investigate?"
    },
    {
      "role": "agent",
      "text": "Exactly. I've also issued you a new card number — your physical replacement card will arrive in 3 to 5 business days. Digital card is available immediately in the app."
    },
    {
      "role": "customer",
      "text": "I can see the new digital card in the app. And I got the fraud case reference: FRD-5521."
    },
    {
      "role": "agent",
      "text": "Perfect. I've also enabled enhanced transaction alerts on your account. Is there anything else I can assist with?"
    },
    {
      "role": "customer",
      "text": "No, thank you. You acted fast and that really reassured me."
    }
  ],
  "resolution": "Unauthorized charges frozen and fraud dispute filed. Provisional credit issued within 24 hours. New card issued digitally and physically."
}

How to Use
Load with Python
python

import json

with open("finance.json", "r", encoding="utf-8") as f:
    data = json.load(f)
    conversations = data["conversations"]

for conv in conversations:
    print(conv["problem"])

Load with Hugging Face Datasets
python

from datasets import load_dataset

dataset = load_dataset("ai-training-datasets/finance", split="train")
print(dataset[0])

Customer Personas Included

    frustrated user

    confused beginner

    impatient executive

    elderly user

    tech-savvy user

    calm and patient user

License

This dataset is released under the MIT License. You are free to use, modify, and distribute it for both research and commercial purposes.
Citation

If you use this dataset, please cite:
text

@dataset{ai_training_datasets_2026,
  title     = {AI Training Datasets: Financial Support Conversations},
  author    = {AI Training Datasets},
  year      = {2026},
  version   = {1.0},
  publisher = {Hugging Face}
}

## 💼 Commercial licensing

This dataset is available for free under the MIT License for non-commercial use.  
If you need a commercial license, custom datasets, or the full version with all 600 conversations, contact me at:

📧 cybernovasg@gmail.com 
Downloads last month
9