Scribd
Upload
8 views

Tp-Andoc2023 24

This document outlines a data mining project aimed at developing predictive models to determine if bank clients will default on loans using various client data. The goals are to follow the CRISP-DM methodology to explore, preprocess, model, and evaluate the data to predict loan defaults and submit a report detailing the process. The work will be assessed based on the quality of analysis, conclusions, and model accuracy.

Uploaded by

mharia
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
8 views

Tp-Andoc2023 24

This document outlines a data mining project aimed at developing predictive models to determine if bank clients will default on loans using various client data. The goals are to follow the CRISP-DM methodology to explore, preprocess, model, and evaluate the data to predict loan defaults and submit a report detailing the process. The work will be assessed based on the quality of analysis, conclusions, and model accuracy.

Uploaded by

mharia
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Análise de Dados e Gestão do Conhecimento

1st Practical Work

Dep. de Eng. Informática 2023/2024 1st semester

The goal of any data mining project is to extract knowledge and patterns from data using a wide range of
methods and techniques. Within big datasets, important relationships can be uncovered that give insights and
can be used to make predictions that impact the business.

This project aims to develop models that are capable of predicting whether or not clients of a bank will default
on a loan. For this purpose, data was collected from several bank clients, containing samples of clients who
defaulted on a loan and clients who didn't.
When customers fail to make timely loan payments, banks incur losses, resulting in annual losses amounting
to millions of rupees. This significantly affects the country's economic growth. In this project, we will analyse
various factors, including the funded amount, location, loan balance, and more, to forecast whether an
individual is likely to default on their loan.
Defaulting on a loan happens when a client misses payments for a specified period of time. When a loan
defaults, it’s sent to a debt collection agency, whose job it is to collect the unpaid funds from the client. The
period between missing a loan payment and having the loan default is known as "delinquency." The
delinquency period helps clients avoid default by giving them extra time to contact the loan servicer and catch
up on missed payments.

Using the algorithms you've learned, the aim is to develop predictive models based on the data provided. The
project must follow the CRISP-DM methodology and include files for its phases in the Python language,
annotated in MarkDown:
• data exploration and preparation;
• data pre-processing;
• creation of models using data mining algorithms;
• evaluation of the models created.

The project must be submitted with a report describing, in as much detail as possible, the process you followed
to obtain your solutions. The report must include the data mining goals, the most relevant data graphical figures

1
Análise de Dados e Gestão do Conhecimento
1st Practical Work

Dep. de Eng. Informática 2023/2024 1st semester

and their interpretation, an explanation of the cleaning and pre-processing of data performed, an
interpretation or evaluation of the models created, and commitments assumed in their development.

The work will be mainly assessed by the quality of the data analysis process, followed by the conclusions
reached, policies or actions proposed, and the accuracy of the models. But more important than the accuracy
of the models is the description of the analysis process and conclusions extracted from the data.

Deadline and submission instructions


- The project should be submitted to Moodle in the discipline area by 24:00 on December 1. From this date,
the note will be penalised by 10%, and no projects will be accepted after the 3th of December.
- The code and report should be placed in a ZIP file with the designation ANDOC-GRPX.zip being X the group
number.
- The presentation and evaluation of the project will be by group and individual on the 12th week (4–8
December).

You might also like