Skip to content

Supervised text classification problem involving real-world complaints received about financial products and services.

Notifications You must be signed in to change notification settings

KennethOdoh/Consumer-Finance-Complaints

Repository files navigation

Consumer Finance Complaints

In this project a consumer complaint database is being investigated to recognize patterns between various variables to help produce models that will help financial organizations make better business decisions that will profit the organization.

We will first look at the background of the dataset to understand the data being worked with. We will then use exploratory data analysis initially to look at any obvious errors in the data to aid with the data preprocessing.

Before any further analysis and exploring of data is completed, we will outline the goal of the project and outline the direction of our investigation by outlining the important questions in the project objectives. After this step is complete, we will use exploratory data analysis to look at the data, gain meaningful insight, understand the structure of the data set and understand the relationship between various variables and note any trends.

Afterwards we will consider the summaries and visualizations to identify relevant business problems and the impact it has on the business. We will then analyze the problems and identify business opportunities which refer to new and innovative ways of doing business that would be made evident by this project.

Through using sophisticated algorithms, some modelling will be performed on a sample of the dataset, and the conclusions from these models will be documented Lastly, there will be some recommendations to these financial institutions regarding new and innovative ways of doing business that would be made possible by the insights made in this project.

About the Dataset

This dataset is real world complaints received about financial products and services. The dataset was provided by the Consumer Financial Protection Bureau. Each complaint is related to a specific product. These complaints represent specific issues, and these occur at specific companies in certain areas.

Due to the large size of the dataset, we could not upload it directly here on GitHub. However, the dataset can be downloaded in either of two ways:

  1. From Kaggle: link
  2. From Google Drive: link

If you choose to download from Kaggle, the csv file should be renamed from 'row.csv'to 'data.csv, and be extracted to the same folder containing the source codes for this project.

Contributors

  1. Kenneth Odoh
  2. David Mbaya Luboya-Mushila
  3. Muskan Gupta
  4. Esuola Stephen Semiloore
  5. Emmanuel Oyetunji
  6. Hawal Alade
  7. Alonge Daniel
  8. Edunoh Inyeneobong Ita-Okon

About

Supervised text classification problem involving real-world complaints received about financial products and services.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages