The Consumer Complaint Database is a collection of complaints about consumer financial products and services that we sent to companies for response. Complaints are published after the company responds, confirming a commercial relationship with the consumer, or after 15 days, whichever comes first. Complaints referred to other regulators, such as complaints about depository institutions with less than $10 billion in assets, are not published in the Consumer Complaint Database. The database generally updates daily.

The Consumer Complaint Database is a collection of complaints about consumer financial products and services that we sent to companies for response. Complaints are published after the company responds, confirming a commercial relationship with the consumer, or after 15 days, whichever comes first. Complaints referred to other regulators, such as complaints about depository institutions with less than $10 billion in assets, are not published in the Consumer Complaint Database.This database is not a statistical sample of consumers’ experiences in the marketplace. Complaints are not necessarily representative of all consumers’ experiences and complaints do not constitute “information” for purposes of the Information Quality Act . Complaint volume should be considered in the context of company size and/or market share. For example, companies with more customers may have more complaints than companies with fewer customers. We encourage you to pair complaint data with public and private datasets for additional context. The Bureau publishes the consumer’s narrative description of his or her experience if the consumer opts to share it publicly and after the Bureau removes personal information. We don’t verify all the allegations in complaint narratives. Unproven allegations in consumer narratives should be regarded as opinion, not fact. We do not adopt the views expressed and make no representation that consumers’ allegations are accurate, clear, complete, or unbiased in substance or presentation. Users should consider what conclusions may be fairly drawn from complaints alone.This public dataset is hosted in Google BigQuery and is included in BigQuery's 1TB/mo of free tier processing. Each user receives 1TB of free BigQuery processing every month, which can be used to run queries on this public dataset. Watch this short video to learn how to get started quickly using BigQuery to access public datasets. What is BigQuery

Each week the CFPB sends thousands of consumers’ complaints about financial products and services to companies for response. Those complaints are published here after the company responds or after 15 days, whichever comes first. By adding their voice, consumers help improve the financial marketplace.

Each week we send thousands of consumers' complaints about financial products and services to companies for response. Complaints are listed in the database after the company responds or after they’ve had the complaint for 15 calendar days, whichever comes first.

We publish the consumer’s description of what happened if the consumer opts to share it and after taking steps to remove personal information. See our Scrubbing Standard for more details

We don’t verify all the facts alleged in these complaints, but we take steps to confirm a commercial relationship. We may remove complaints if they don’t meet all of the publication criteria. Data is generally refreshed nightly. Company level information should be considered in context of company size and/or market share.

More about the Consumer Complaint Database | How we use complaint data | Technical documentation

Each week we send thousands of consumers' complaints about financial products and services to companies for response. Those complaints are published here after the company responds, confirming a commercial relationship with the consumer, or after 15 days, whichever comes first. Complaint narratives are consumers' descriptions of their experiences in their own words. By adding their voice, consumers help improve the financial marketplace. The database generally updates daily.

These are complaints we’ve received about financial products and services.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F16731800%2F68da45b26ffa5db1a68a4e1589eb8fba%2Fgraph1.gif?generation=1740945615150985&alt=media" alt=""> https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F16731800%2Fec069b191ef6c0c36d24f129e9fecc80%2Ffoto3.png?generation=1740945620897811&alt=media" alt=""> https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F16731800%2F7e45c8e94dd53b5d9332be8185c67d1c%2Fgraph2.gif?generation=1740945628087024&alt=media" alt="">

When people have problems with financial products and services, they can file a complaint with the Consumer Financial Protection Bureau (CFPB). This agency collects complaints and sends them to the company involved to help solve the issue.

Between 2017 and 2023, many customers filed complaints about Bank of America related to different financial products, such as bank accounts, credit cards, loans, and mortgages. Each complaint includes details like:

The date it was submitted to the CFPB. The date the CFPB sent it to the bank for review. The specific financial product involved (e.g., checking account, credit card, mortgage). The issue (e.g., unauthorized transactions, loan repayment problems, fees). The bank's response (e.g., refunding money, explaining the issue, or rejecting the complaint). Common Issues in Complaints Unauthorized Transactions – Customers reported money missing from their accounts or transactions they didn’t make. High or Unexpected Fees – Some people were charged fees they didn’t expect, like overdraft or maintenance fees. Loan and Mortgage Problems – Customers faced issues with loan payments, refinancing, or incorrect charges. Credit Card Disputes – Some users had trouble resolving incorrect charges on their credit cards. How Bank of America Responded The bank usually responded by:

Providing an explanation for the charge or issue. Issuing refunds when mistakes were found. Denying the complaint if they believed no error occurred. Not all complaints resulted in a solution for the customer, but reporting issues helps the CFPB track banking problems and ensure companies follow fair financial practices.

These are complaints received to the Consumer Financial Protection Bureau about financial products and services in Utah.

Problem Statement

When consumers are not happy with some aspect of a business, they choose to reach out to the customer service and might raise a complaint. Businesses try their best to resolve the complaints that they receive. However, it might not always be possible to appease every customer.

Unhappy consumers might raise follow-up questions/complaints about the resolutions provided, and this is detrimental to the business as it points to systemic failures in the Customer Support division and could lead to poor brand image. Disputed complaints which are being/have been resolved could be a critical dataset to derive essential learnings for any business.

Predicting whether a complaint resolution will be accepted or rejected by a consumer can enable a business to proactively look at complaints which might be disputed and hence save unnecessary escalation as well as their reputation. Systemic issues can be identified by noticing which complaints have a higher potential to be disputed, and customer support agents can be trained to pay more attention or enhance the quality of communication for certain types of complaints.

The Consumer Financial Protection Bureau (CFPB) in the United States receives several consumers’ complaints about the dealings of financial companies. It sends these complaints about their products and services to them for eliciting a response. The CFPB makes sure that these complaints are published here soon after the company responds or after 15 days since sending the complaint to the company.

Dataset

You have been provided with a dataset containing the following columns –

● Date received: Date when the complaint was received

● Product: Type of product identified in the complaint, e.g., “Student loan”

● Sub-product: Type of sub-product identified in the complain

● Issue: The issue raised in the complaint, e.g., “Struggling to repay your loan.”

● Sub-issue: E.g., “Problem lowering your monthly payments.”

● Consumer complaint narrative: This is a consumer-submitted description of “what happened”. Reasonable steps have been taken to remove personal information that could be used to identify the consumer

● Company public response: The response to a consumer’s complaint. It can be from a pre-set list of options, e.g., “Company believes the complaint is the result of an isolated error”

● Company: For which the complaint is about

● State: Derived from the consumer’s mailing address

● ZIP Code: Derived from the consumer’s mailing address

● Consumer consent provided: Flag to specify whether the consumer allowed the publishing of their complaint description

● Submitted via: E.g., “Web” or “Phone.”

● Date sent to the company

● Company response to consumer

● Timely response: Flag specifying if the response was timely

● Consumer disputed: Flag specifying if the consumer disputed the resolution

● Complaint ID: Identifier for each complaint

Two files have been provided.

● Training Data: Consumer_Complaints_train.csv

● Test Data: Consumer_Complaints_test.csv

This data set contains complaints by Utah citizens to the Consumer Financial Protection Bureau. These complaints are published here after the company responds or after 15 days, whichever comes first. By adding their voice, consumers help improve the financial marketplace.

(PENTING)

Dataset yang digunakan dalam tantangan ini merupakan versi ringan dan terkurasi dari Consumer Complaint Database, yang dikelola oleh Consumer Financial Protection Bureau (CFPB); lembaga independen di Amerika Serikat yang menangani perlindungan konsumen di sektor keuangan.

Setiap entri dalam dataset mencerminkan keluhan nyata dari konsumen terhadap produk dan layanan keuangan, lengkap dengan narasi bebas, label produk dan isu utama, tanggapan perusahaan, serta informasi tambahan lainnya. Data ini memuat berbagai informasi penting yang dapat dimanfaatkan untuk memahami pola isu yang sering terjadi dan meningkatkan kualitas layanan keuangan.

Untuk mempermudah proses eksplorasi dan khususnya modeling, versi dataset yang disediakan telah disampling secara harian dengan bobot tertentu, sehingga menghasilkan distribusi seimbang antara data teks narasi (tidak missing) dan data tanpa teks narasi. Tujuannya adalah untuk menjaga keberagaman data sambil tetap mempertimbangkan keterbatasan sumber daya komputasi peserta.

Peserta diperbolehkan menggunakan dataset asli CFPB apabila menginginkan cakupan data yang lebih luas. Penggunaan dataset tambahan dari sumber lain juga diperbolehkan, selama peserta mencantumkan sumbernya dengan jelas dan menjelaskan relevansinya terhadap solusi yang diajukan.