File size: 2,685 Bytes
da84941 219bc3d e1b1382 cf40548 2917d59 cf40548 2917d59 cf40548 5ee6f87 cf40548 1e8ff40 134a694 1e8ff40 134a694 030c1b4 134a694 030c1b4 134a694 6ab048f 134a694 6ab048f 134a694 6ab048f 134a694 6ab048f 134a694 6ab048f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 |
---
license: mit
datasets:
- atulgupta002/banking_customer_service_query_intent
language:
- en
base_model:
- google-bert/bert-base-uncased
pipeline_tag: text-classification
tags:
- finance
- banking
- intent
- classification
- customer
- service
- BERT
---
# Banking Customer Service Intent Classifier
This model is designed to classify customer service queries into different intents, based on the type of inquiry made by the customer. It was fine-tuned on a **synthetic** dataset of realistic banking customer service interactions and can classify the following intents:
- `transaction_query`
- `password_reset`
- `loan_inquiry`
- `fraud_report`
- `credit_card_application`
- `balance_inquiry`
## Dataset
Link: https://huggingface.co/datasets/atulgupta002/banking_customer_service_query_intent
## Model Overview
The model is a fine-tuned BERT-based architecture that classifies text inputs into one of the six specified intents. It leverages the **transformers** library by Hugging Face for tokenization and model loading.
## Intended Use
This model is suitable for deployment in applications that require automatic classification of customer service queries, such as:
- Chatbots
- Virtual assistants
- Automatic re-routing incoming emails,calls, and texts
It can be used to classify various types of banking queries, such as requests for account balance, loan inquiries, or fraud reports.
## Installation
To install the necessary dependencies, use the following:
```bash
pip install transformers torch
```
## Inference
```bash
from transformers import AutoModelForSequenceClassification, AutoTokenizer
import torch
labels = [
'transaction_query',
'password_reset',
'loan_inquiry',
'fraud_report',
'credi_card_application',
'balance_inquiry'
]
label2id = {label: idx for idx, label in enumerate(labels)}
id2label = {idx: label for label, idx in label2id.items()}
# Load the pre-trained model and tokenizer
model = AutoModelForSequenceClassification.from_pretrained("atulgupta002/banking_customer_service_query_intent_classifier")
tokenizer = AutoTokenizer.from_pretrained("atulgupta002/banking_customer_service_query_intent_classifier")
def predict(text):
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
with torch.no_grad():
outputs = model(**inputs)
logits = outputs.logits
predicted_class_id = logits.argmax().item()
return id2label[predicted_class_id]
query = "I want to apply for a new credit card"
print(predict(query))
```
## Sample output
|