Project Title

Bank Risk Controller Systems

Skills take away From Python, Analytics, statistics, Ploting, streamlit ,

This Project machine learning, Deep Learning, GenAI

Domain Banking

Problem Statement:

Please refer the given data and do the prediction of whether the customer will
get default or not. The column which you are going to predict in dataset is
“TARGET”.

Business Use Cases:

Potential Business Scenarios

Risk Management: Banks and financial institutions can use the model to
assess the risk of potential borrowers defaulting on loans, helping in decision-
making for loan approvals.
Customer Segmentation: Segment customers based on their risk profile to
tailor financial products and services.
Credit Scoring: Enhance traditional credit scoring models with predictive
analytics to improve accuracy.
Fraud Detection: Identify patterns that may indicate fraudulent loan
applications.

Approach:

Methodology
Data Collection: Gather historical loan data from various sources.
Data Preprocessing: Clean and preprocess the data to handle missing values,
outliers, and categorical variables.
Exploratory Data Analysis (EDA): Perform EDA to understand data
distributions and relationships between variables.
Feature Engineering: Create new features that could enhance the predictive
power of the model.
Model Selection: Compare various machine learning models such as Logistic
Regression, Decision Trees, Random Forest, and Gradient Boosting, etc. to
find the best fit.
 Model Training: Train the selected model(s) on the training dataset.
Model Evaluation: Evaluate the model performance using appropriate metrics.
Hyperparameter Tuning: Optimize the model parameters to improve
performance.

Model Deployment: Deploy the model for real-time prediction and integrate it
with business systems.

Results:

Expected Outcomes:
Create a streamlit application dashboard with multiple sidebar menus and do the
following operations in respective sidebars.
Sidebar Sidebar Name Function to do
Number
1 Data 1. Show the dataset which you
are using for model building.
2. Show the model
performance metric dataset.
2 EDA - Visual 1. Do all kind of EDA analysis
for the given dataset.
2. Plot the EDA plots and
Statistics plots and show
them in this menu.
3. Plot preferred are Plotly or
seaborn, try to plot in plotly
for interaction.
3 Prediction 1. Create a text or number
input items in streamlit for
selected featured which you
have selected for model
building.
2. Once the user has entered
the value and hit the button
of predict, you model should
predict and show whether
the customer is defaulted or
not.
4 [optional] NLP 1. Create a text box and get
the text from user and do
the NLP preprocessing
techniques.
 2. Provide the sentiment
analysis plot for the text
given.
3. Use RNN or LSTM model for
this (free to use your own
datset),
4. Predict the upcoming text.
5 [optional] Recommendation 1. Collect your own data for
recommendation system.
2. You can use your any of the
data to build
recommendation it may be
a product recommendation,
or service recommendation.
6 [optional] Deep Learning Detect only humans from the
[Object detection] image using Yolov5 model.
(You can use any pretrained
model also)

7 [optional] Chat - GenAI Create a chatbot to resolve the

bank customer queries.
Bot have to respond with
answers to the question.
Use the bank pdf documents as
input. (bank statement,
policy document, etc)

A trained machine learning model capable of predicting loan defaulters with high
accuracy.

Project Evaluation metrics:

Success Criteria
Accuracy, Precision, Recall, F1 Score, ROC-AUC
All should be in high range, minimum 0.87 or 87 %.

Technical Tags:

Machine Learning
Data Preprocessing
Feature Engineering
Model Training
Model Evaluation
Hyperparameter Tuning
Deep Learning
GenerativeAI
Llama2
GPT

Deployment

Data Set:

Dataset is available in CSV format.

Datas Loan_data.cs [Link]
et v oTkj3Gla0H/view?usp=sharing

Data Data_dictiona [Link]

Diction [Link] PL8Ks_fs6SI/view?usp=sharing
ary

Data Set Explanation:

Content and Context

Historical Data: The dataset contains information about past loan applications,
including personal details of applicants and their loan status.
Preprocessing Steps:
Handle missing values by imputation.
Encode categorical variables using techniques like one-hot encoding.
Normalize or standardize numerical variables if required.
Create new features based on domain knowledge to improve model performance.

Project Deliverables:

Submission Requirements
Source Code: The complete code used for data preprocessing, model training, and
evaluation.
Documentation: A report detailing the methodology, analysis, results, and insights.
Presentation: A slide deck summarizing the project and key findings.
Model Files: The trained model ready for deployment.
README: Instructions on how to run the code and reproduce the results.

Project Guidelines:

Best Practices
Coding Standards: Standard code standard for Python code.
Version Control: Use Git for version control and regularly commit changes.
Documentation: Comment your code and provide clear explanations for your logic.
Collaboration: Use collaborative tools like GitHub or GitLab for team projects.

Timeline:

Analyse data
EDA
Ploting 6 Days
Building Model
ML Model Selection

NLP (Optional) 5 Days

Recommendation system (Optional) 5 Days

Deep Learning (Optional) 5 Days
GenAI (Optional) 5 days
Streamlit 3 Days
Total 2 weeks

PROJECT DOUBT CLARIFICATION SESSION ( PROJECT AND CLASS DOUBTS)

About Session: The Project Doubt Clarification Session is a helpful resource for resolving questions
and concerns about projects and class topics. It provides support in understanding project requirements,
addressing code issues, and clarifying class concepts. The session aims to enhance comprehension and
provide guidance to overcome challenges effectively.
Note: Book the slot at least before 12:00 Pm on the same day

Timing: Tuesday, Thursday, Saturday (5:00PM to 7:00PM)

Booking link :[Link]

LIVE EVALUATION SESSION (CAPSTONE AND FINAL PROJECT)

About Session: The Live Evaluation Session for Capstone and Final Projects allows participants to
showcase their projects and receive real-time feedback for improvement. It assesses project quality and
provides an opportunity for discussion and evaluation.
Note: This form will Open on Saturday and Sunday Only on Every Week

Timing: Monday-Saturday (11:30PM to 12:30PM)

Booking link : [Link]