The banking system has a lot of factors that can be improved by the use of Machine Learning and Artificial Intelligence. This is a proposed project that will use certain ML algorithms to make the system smarter.
2. Project Guide : Ms. Lino Murali
Project Co-Ordinator : Mr. Pramod Pavithran
Head Of the Department : Mr. V. Damodaran
3. 1. Overview
The field of banking is one of the most important ones in the financial
sectors. The need for automation of such a sector is a needful deed as
the usage is increasing day by day and to a great extent.
There is a scope of improvement in a lot of areas of the banking
sector. Right from the perspective of the bank users to the bank
themselves, the introduction of machine learning based techniques
can be a rightful save.
4. 2. Introduction
● New customer channels are changing the face of traditional
banks and disrupting existing banking models. Rising mobile
penetration has transformed the way consumers bank.
● The use of ML, AI in most of the banks in India still in idea phase
● SwedBank’s chatbot handles around 40k customer calls/month
● Singapore Headquartered DBS bank launched a banking app in
India last year with built in AI.
● Banks like ICICI, HDFC,SBI are in pilot/testing phase.
● Applications of ML,AI implemented in banks of India- fraud
detection, chatbots.
6. 5. Problem Statement
1. Loan Approval Prediction : Predicting whether a loan applied by
a customer will be approved or not.
2. Recommendation Engine : Recommending the customers about
best savings strategies, loan selection plans and insurance
options etc.
3. Twitter Sentiment Analysis: Analyzing sentiments and feedback
of users by extracting data from social media platforms and
gaining insights on areas where the bank needs to improve.
7. 6. Existing System
The current system of banking in India does not use Machine
Learning and Artificial Intelligence applications. The banks only are in
idea phase or testing phase. The only applications of AI that have
been implemented are- chatbots(HDFC) and fraud detection. But
there are many other sectors that lack the implementation and
innovation in finance and banking that can be done with wide
varieties of ML application.
9. 7. Proposed System
● We come up with a set of solution to make a software that
makes the banking system smart through applications of ML,AI.
● This system will be able to predict the loan approval status
beforehand so as to make a hassle free situation for its
customers. As well as, the banks will know well in advance who
can be the defaulter and hence help in decision making.
● Banks will be able to recommend the users about its latest
schemes/products launched to its target audience. This will
increase the efficiency and reduce cost as well as efforts.
● The sentiments/feedbacks/complaints of customers will be
analyzed through social media and various areas can be
improved upon in the banks based on the insights.
12. Anaconda
1. Anaconda is a python distribution, with installation and package
management tools.
2. It provides large selection of packages and commercial support.
It is an environment manager which provides the facility to
create different python environment, each with their own
setting.
3. It also provides much greater advantages in data science
platform
13. NumPy
● NumPy is a python library for linear algebra. At its core is the
NumPy Array, a multidimensional data structure that can be
used to represent vector and matrices
● NumPy gives us much of the functionality we’d have in a
scientific computing language like Matlab, R.
● It forms the backbone of python’s very popular scientific
Computing stack, which is used a lot in data science, ML.
14. Pandas
● It is an open source, BSD-licensed library providing high-
performance, easy-to-use data structures and data analysis tools
for Python.
18. 11. Sentiment Analysis
● Sentiment analysis refers to the processes, methods, techniques,
and approaches that retrieve information about consumer
attitude toward a product, service, or brand.
● Also known as Opinion Mining or Emotion AI.
25. 13. Conclusion
Our proposed system will hence bring a change in the existing system
of banks. The banks will be smart enough to predict the loan approval
well in advance that will benefit the users as well as bank. The banks
will be able to recommend its target users about its latest schemes
launched.