daiha98 / credit_scoring

RMarkdowns and HTML output files about credit risk and score in loan request process. Includes Data Analysis (DA) & Machine Learning (ML) codes. For deployment, the API was built using Plumber and Docker to host my application.

README

1. INTRODUCTION:

• This file contains the informations about a dataset that is related to credit risk & score. It's a tiny dataset with about 40000 rows and 29 variables that the goal is to work with data analysis/visualization and machine learning algorithms to predict the target BAD.

2. PROGRAMMING LANGUAGE:

• This code was made on R, which is a language that is not entirely unlike (versions 3 and 4 of) the S language developed at AT&T Bell Laboratories by Rick Becker, John Chambers and Allan Wilks. R is free software distributed under a GNU-style copyleft. The core of R is an interpreted computer language with a syntax superficially similar to C, but which is actually a "functional programming language" with capabilities similar to Scheme. The language allows branching and looping as well as modular programming using functions.

3. ABOUT THE DATASET:

• The 'CreditScoring' data frame dim contains 40000 rows and 29 variables:

  • V1 = Index counter;
  • clientId = Client index;
  • shopId = Loan index;
  • gender = Client gender;
    • 'F' = Feminino
    • 'M' = Masculino
  • maritalStatus = Client marital status;
    • 'C' = Casado
    • 'D' = Divorciado
    • 'S' = Solteiro
    • 'V' = Viúvo
    • 'O' = Outros
  • age = Client age;
  • numberOfDependents = Client number of dependents;
  • education = Client schooling;
  • flagResidencialPhone: Flag if client has declared residencial phone;
    • Y = Sim
    • N = Não
  • areaCodeResidencialPhone: Area code from declared residencial phone;
  • bestPaymentDay: Choosen day client has declared as best day to pay a loan;
  • shopRank: Loan rank on plataform;
  • residenceType: Client residence type;
    • 'P' = Própria
    • 'A' = Alugada'
    • 'C' = Compartilhada
    • 'O' = Outros
  • monthsInResidence: Number of months client is living in declared residence;
  • flagMothersName: Flag if client has declared mothers name;
    • Y = Sim
    • N = Não
  • flagFathersName: Flag if client has declared fathers name;
    • Y = Sim
    • N = Não
  • flagResidenceTown_WorkingTown: Flag if client live in his/her working town;
    • Y = Sim
    • N = Não
  • flagResidenceState_WorkingState: Flag if client live in his/her working state;
    • Y = Sim
    • N = Não
  • monthsInTheJob: Number of months client is working in his/her actual job;
  • professionCode: Client profession code
  • mateIncome: Client mate income
  • flagResidencialAddress_PostalAddress: Flag validation zip code client;
    • Y = Sim
    • N = Não
  • flagOtherCard: Flag if client has others credit card;
    • Y = Sim
    • N = Não
  • quantityBankAccounts: Client number of bank accounts;
  • flagMobilePhone: Flag if client has mobile phone;
    • Y = Sim
    • N = Não
  • flagContactPhone: Flag if client has contact phone;
    • Y = Sim
    • N = Não
  • personalNetIncome: Client personal income;
  • codeApplicationBooth: Code where client requested his/her loan application;
  • BAD: Flag if client has delayed his/her loan
    • 0 = Adimplente
    • 1 = Inadimplente