This repository contains a collection of machine learning tasks and analyses performed on a dataset of consumer complaints about business-related issues in Brazil. The dataset covers various aspects of complaints, including their subjects, regions, and more.
The dataset used for these analyses can be found in the data
folder. It includes the following columns:
- AnoCalendario
- DataArquivamento
- DataAbertura
- CodigoRegiao
- Regiao
- UF
- strRazaoSocial
- strNomeFantasia
- Tipo
- NumeroCNPJ
- RadicalCNPJ
- RazaoSocialRFB
- NomeFantasiaRFB
- CNAEPrincipal
- DescCNAEPrincipal
- Atendida
- CodigoAssunto
- DescricaoAssunto
- CodigoProblema
- DescricaoProblema
- SexoConsumidor
- FaixaEtariaConsumidor
- CEPConsumidor
This repository provides implementation for various machine learning tasks on the dataset, including:
- Binary Classification - Complaint Resolution Prediction (Solved)
- Multi-Class Classification - Issue Category Prediction (Solved)
- Regression - Time to Resolution Prediction
- Clustering - Regional Complaint Patterns
- Anomaly Detection - Unusual Complaints
- Natural Language Processing (NLP) - Sentiment Analysis
- Feature Importance Analysis
For each task, you will find Jupyter Notebook files (*.ipynb
) that walk you through the process of data preprocessing, model training, evaluation, and analysis.
-
Clone this repository to your local machine using your preferred method.
git clone https://github.com/Dank-del/brazil-consumer-complaints-analysis.git
-
Set up your environment using Anaconda (recommended):
-
Install Anaconda.
-
Create a new Anaconda environment using the provided
environment.yml
file:conda env create -f environment.yml
-
Activate the new environment:
conda activate consumer-complaints
-
-
Launch Jupyter Notebook and navigate to the specific task folder:
jupyter notebook
-
Open the corresponding Jupyter Notebook (
*.ipynb
) for the task you're interested in. -
Follow the instructions in the notebook to run the analysis and understand the results.
This project is licensed under the MIT License - see the LICENSE file for details.