Update README.md
Browse files
README.md
CHANGED
|
@@ -4,6 +4,7 @@ tags:
|
|
| 4 |
- sklearn
|
| 5 |
- skops
|
| 6 |
- tabular-classification
|
|
|
|
| 7 |
model_format: pickle
|
| 8 |
model_file: skops-ise057qg.pkl
|
| 9 |
widget:
|
|
@@ -27,16 +28,22 @@ widget:
|
|
| 27 |
---
|
| 28 |
|
| 29 |
# Model description
|
| 30 |
-
|
| 31 |
-
|
| 32 |
|
| 33 |
## Intended uses & limitations
|
| 34 |
-
|
| 35 |
-
[More Information Needed]
|
| 36 |
|
| 37 |
## Training Procedure
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 38 |
|
| 39 |
-
[
|
| 40 |
|
| 41 |
### Hyperparameters
|
| 42 |
|
|
@@ -97,26 +104,7 @@ widget:
|
|
| 97 |
|
| 98 |
|
| 99 |
|
| 100 |
-
# How to Get Started with the Model
|
| 101 |
-
|
| 102 |
-
[More Information Needed]
|
| 103 |
|
| 104 |
# Model Card Authors
|
| 105 |
|
| 106 |
-
This model card is written by following authors:
|
| 107 |
-
|
| 108 |
-
[More Information Needed]
|
| 109 |
-
|
| 110 |
-
# Model Card Contact
|
| 111 |
-
|
| 112 |
-
You can contact the model card authors through following channels:
|
| 113 |
-
[More Information Needed]
|
| 114 |
-
|
| 115 |
-
# Citation
|
| 116 |
-
|
| 117 |
-
Below you can find information related to citation.
|
| 118 |
-
|
| 119 |
-
**BibTeX:**
|
| 120 |
-
```
|
| 121 |
-
[More Information Needed]
|
| 122 |
-
```
|
|
|
|
| 4 |
- sklearn
|
| 5 |
- skops
|
| 6 |
- tabular-classification
|
| 7 |
+
- finance
|
| 8 |
model_format: pickle
|
| 9 |
model_file: skops-ise057qg.pkl
|
| 10 |
widget:
|
|
|
|
| 28 |
---
|
| 29 |
|
| 30 |
# Model description
|
| 31 |
+
This is a Gaussian Naive Bayes model trained on a synthetic dataset, containining a large variety of transaction types representing normal activities as well as
|
| 32 |
+
abnormal/fraudulent activities generated by J.P. Morgan AI Research. The model predicts whether a transaction is normal or fraudulent.
|
| 33 |
|
| 34 |
## Intended uses & limitations
|
| 35 |
+
Terms of use for the J.P. Morgan AI Research synthetic dataset limits sharing of the dataset
|
|
|
|
| 36 |
|
| 37 |
## Training Procedure
|
| 38 |
+
The data preprocessing steps applied include the following:
|
| 39 |
+
- Dropping high cardinality features or no variance features. This include Transaction ID, Sender ID, Sender Account, Beneficiary ID, Beneficiary Account, Sender LOB, Sender Sector and Time
|
| 40 |
+
- Transforming and Encoding categorical features namely: Sender Country, Beneficiary Country, Transaction Type, and the target variable, Label
|
| 41 |
+
- Applying feature scaling on all features
|
| 42 |
+
- Splitting the dataset into training/test set using 85/15 split ratio
|
| 43 |
+
- Handling imbalanced dataset using imblearn framework and applying RandomUnderSampler method to eliminate noise
|
| 44 |
+
|
| 45 |
|
| 46 |
+
|
| 47 |
|
| 48 |
### Hyperparameters
|
| 49 |
|
|
|
|
| 104 |
|
| 105 |
|
| 106 |
|
|
|
|
|
|
|
|
|
|
| 107 |
|
| 108 |
# Model Card Authors
|
| 109 |
|
| 110 |
+
This model card is written by following authors: Seifullah Bello
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|