Scribd
Upload
7 views1 page

Banking System Profit Maximization Report

The project aimed to maximize profits for a banking system using a software-based model. The team has completed 90% of the work, including pre-processing a dataset of 1000 entries with 20 attributes, exploring the data, applying machine learning algorithms for classification, building models, and comparing different algorithms. They analyzed loan applications and assessed the accuracy of classification models for "bad" and "good" groups when changing thresholds. The main difficulty was assessing hard-to-measure criteria like borrowers' status and capacity that could lead to errors in lending decisions.

Uploaded by

Swarnima Singh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
7 views1 page

Banking System Profit Maximization Report

The project aimed to maximize profits for a banking system using a software-based model. The team has completed 90% of the work, including pre-processing a dataset of 1000 entries with 20 attributes, exploring the data, applying machine learning algorithms for classification, building models, and comparing different algorithms. They analyzed loan applications and assessed the accuracy of classification models for "bad" and "good" groups when changing thresholds. The main difficulty was assessing hard-to-measure criteria like borrowers' status and capacity that could lead to errors in lending decisions.

Uploaded by

Swarnima Singh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Project Progress Report-III

Title Of Project: Profit Maximisation For Banking System

Type of Model: Software based

Project Objective:

This can be achieved in following steps:

- Data pre processing


- Exploratory Data analytics: Analysis of socio economic attributes.
- Application of various algorithms, each of which presented with a short introduction
- Model Building
- Choice of model
- Comparison between different forms of algorithms, if more than one is used.
- Result Analysis

We have achieved our six objectives that is data pre processing, exploratory data analysis, Application
of various algorithms, model building, choice of model and comparision between different algorithm.

Percentage Of Work Completed: 90% of the work have been completed.

Progress Of Work:

 In data prep processing, dataset is pre processed and cleaned for achieving better result in
applied model in machine [Link] dataset that is used in this project is
”Nordic_kredit.csv”.It consist of 1000 entries with 20 attributes.
 Using four machine learning models for the problem classification. For each of the
classification thresholds examine the variability of three criteria for evaluating the
quality of classification.
 Assessing the quality of classification of two groups of loan applications(bad and
good groups) as well as general accuracy level of classification models when the
threshold changes.

Difficulties: The basis for granting credit to a customer is based on a number of criteria that some of
them are difficult to measure such as status and capacity of borrower are difficult to assess and may
therefore lead to errors in lending decisions.

Project Guide: Mrs Nidhi Saxena


Name of team members:
Swarnima Singh(1712210909)
Divyanshi Singh(1712210905)

You might also like