Text Classification
Transformers
Safetensors
English
distilbert
bank user complaint
intent-classifier
Instructions to use KunalEsM/bank_complaint_intent_classifier_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use KunalEsM/bank_complaint_intent_classifier_v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="KunalEsM/bank_complaint_intent_classifier_v2")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("KunalEsM/bank_complaint_intent_classifier_v2") model = AutoModelForSequenceClassification.from_pretrained("KunalEsM/bank_complaint_intent_classifier_v2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
library_name: transformers
|
| 8 |
+
pipeline_tag: text-classification
|
| 9 |
+
tags:
|
| 10 |
+
- bank user complaint
|
| 11 |
+
- intent-classifier
|
| 12 |
+
|
| 13 |
+
widget:
|
| 14 |
+
- text: "I was charged Rs. 500 for an SMS alert that I did not request. Can you please reverse this charge and deactivate the service?"
|
| 15 |
+
example_title: "Example- Dispute in charges deducted"
|
| 16 |
+
- text: "My account was levied with a Rs. 436 minimum balance charge on 15th June 2022. Please explain the reason for this deduction."
|
| 17 |
+
example_title: "Example- Minimum Balance Charges Related"
|
| 18 |
+
- text: "On 8th June 2023, I was charged Rs. 200 for an RTGS transaction that was not authorized by me. Kindly investigate and reverse this charge."
|
| 19 |
+
example_title: "Example- Fund Remittance: NEFT/ RTGS/ IMPS through Branch"
|
| 20 |
+
- text: "I received a charge of Rs. 1000 for my education loan, but I have already made a full payment. Can you please explain why this charge was applied?"
|
| 21 |
+
example_title: "Example- Education Loans"
|
| 22 |
+
- text: "I have noticed a discrepancy in the processing fees for my State Government Scheme loan. On 15th December 2023"
|
| 23 |
+
example_title: "Example- Govt Scheme Loan"
|
| 24 |
+
- text: "my Home Loan account was debited with Rs. 1500 without any explanation."
|
| 25 |
+
example_title: "Example- Home Loan"
|
| 26 |
+
- text: "I was surprised to see a charge of Rs. 2569 on my Other Advance account. Can you please clarify what this charge is for?"
|
| 27 |
+
example_title: "Example- Other Advance amount"
|
| 28 |
+
- text: "I have been charged Rs. 9000 as a processing fee, but I was informed that there would be no processing fees for my loan."
|
| 29 |
+
example_title: "Example- Personal Loan"
|
| 30 |
+
- text: "Our Cash Credit (CC) Limit account with your bank was renewed in December 2023. However, we have been charged twice for the account renewal amount of Rs. 2500. Requesting an immediate reversal of the excess charge."
|
| 31 |
+
example_title: "Example- SME Advances"
|
| 32 |
+
- text: "I was charged a documentation fee of Rs. 3500 for my Vehicle Loan, which was not approved due to incomplete documentation. As per the loan terms, this fee should be waived. I request a refund for this amount."
|
| 33 |
+
example_title: "Example- Vehicle Loan"
|
| 34 |
+
- text: "I received a notification stating that I will not be charged for ATM AMC as long as I maintain a minimum balance. However, on 20/09/2022, I was charged Rs. 250 for ATM annual maintenance. Please clarify why I was charged and provide a refund if necessary."
|
| 35 |
+
example_title: "Example- ATM Related"
|
| 36 |
+
- text: "I received a credit card statement with charges that I did not make. Can you help me dispute these charges and get them reversed?"
|
| 37 |
+
example_title: "Example- Others"
|
| 38 |
+
---
|
| 39 |
+
|
| 40 |
+
# Bank User Query Intent Classification Model
|
| 41 |
+
|
| 42 |
+
This model classifies queries into twelve distinct buckets:
|
| 43 |
+
|
| 44 |
+
- Sample Label: Category, Type, Subtype
|
| 45 |
+
- **Label 0**: DEPOSIT, SB/ CA/TERM DEPOSIT ACCOUNTS, Dispute in charges deducted
|
| 46 |
+
- **Label 1**: DEPOSIT, SB/ CA/TERM DEPOSIT ACCOUNTS, Minimum Balance Charges related
|
| 47 |
+
- **Label 2**: DIGITAL BANKING, ATM RELATED, Dispute in ATM AMC Charges
|
| 48 |
+
- **Label 3**: DIGITAL BANKING,FUND REMITTANCE: NEFT/ RTGS/ IMPS through Branch,Dispute in charges deducted
|
| 49 |
+
- **Label 4**: LOANS & ADVANCES, Education Loans, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 50 |
+
- **Label 5**: LOANS & ADVANCES, Govt Scheme loans, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 51 |
+
- **Label 6**: LOANS & ADVANCES, Home loans, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 52 |
+
- **Label 7**: LOANS & ADVANCES, OTHER ADVANCES, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 53 |
+
- **Label 8**: LOANS & ADVANCES, Personal loans, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 54 |
+
- **Label 9**: LOANS & ADVANCES, SME ADVANCES, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 55 |
+
- **Label 10**: LOANS & ADVANCES, VEHICLE LOANS, Discrepancy in Charges (Processing Fee/Documentation charges, Inspection charges, etc)
|
| 56 |
+
- **Label 11**: Others
|
| 57 |
+
|
| 58 |
+
## Training Metrics
|
| 59 |
+
|
| 60 |
+
The following training metrics were observed over 10 epochs:
|
| 61 |
+
|
| 62 |
+
| Epoch | Loss | Accuracy | F1 Score |
|
| 63 |
+
|-------|---------|----------|----------|
|
| 64 |
+
| 1 | 0.4103 | 0.9182 | 0.9184 |
|
| 65 |
+
| 2 | 0.0672 | 0.9827 | 0.9828 |
|
| 66 |
+
| 3 | 0.0351 | 0.9917 | 0.9917 |
|
| 67 |
+
| 4 | 0.0221 | 0.9948 | 0.9948 |
|
| 68 |
+
| 5 | 0.0171 | 0.9942 | 0.9943 |
|
| 69 |
+
| 6 | 0.0107 | 0.9966 | 0.9966 |
|
| 70 |
+
| 7 | 0.0056 | 0.9989 | 0.9989 |
|
| 71 |
+
| 8 | 0.0037 | 0.9986 | 0.9986 |
|
| 72 |
+
| 9 | 0.0152 | 0.9955 | 0.9955 |
|
| 73 |
+
|10 | 0.0061 | 0.9982 | 0.9982 |
|
| 74 |
+
|
| 75 |
+
|