Dark Mode
Client Credit Status Classification Dataset
Finance & Banking Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset is designed for credit card approval prediction, framing it as a binary classification challenge. It provides a robust set of features related to client demographics, financial status, and contact information, alongside a label indicating whether an application was approved or rejected. It is ideal for developing and evaluating machine learning models aimed at automating and improving credit assessment processes.
-
Columns
-
Ind_ID: Unique identifier for each client. This also serves as the joining key to the label dataset.
-
Gender: Information on the client's gender (Male/Female).
-
Car_owner: Indicates whether the client owns a car (True/False).
-
Propert_owner: Indicates whether the client owns property (True/False).
-
Children: The number of children the client has.
-
Annual_income: The client's annual income.
-
Type_Income: The category of the client's income, e.g., Working, Commercial associate.
-
Education: The client's highest education level, e.g., Secondary / secondary special, Higher education.
-
Marital_status: The client's marital status, e.g., Married, Single / not married.
-
Housing_type: The client's living style, e.g., House / apartment, With parents.
-
Birthday_count: The client's age, represented as a backward count from the current day (0 for today, -1 for yesterday).
-
Employed_days: The start date of employment, also represented as a backward count from the current day. A positive value indicates unemployment.
-
Mobile_phone: Indicates if the client has a mobile phone.
-
Work_phone: Indicates if the client has a work phone.
-
Phone: Indicates if the client has any phone number.
-
EMAIL_ID: Indicates if the client has an email address.
-
Type_Occupation: The client's occupation.
-
Family_Members: The total number of family members.
-
Label: The target variable, where 0 signifies an approved application and 1 signifies a rejected application.
-
Distribution
The dataset consists of two primary files:
Credit_Card.csv containing applicant details and Credit_card_label.csv with the approval status. The main applicant data file is approximately 186.82 KB and features 18 columns.
The dataset contains 1,548 records (rows) with detailed information for each client.
Key data distributions include:-
Gender: 63% Female, 37% Male, with 7 missing entries.
-
Car Ownership: 40% own a car, 60% do not.
-
Property Ownership: 65% own property, 35% do not.
-
Children: The majority of clients have 0 children (1,396 records), with a maximum of 14 children observed. The average is 0.41 children.
-
Annual Income: Ranges from £33,750 to £1,575,000, with a mean income around £191,000. 1% of entries are missing for this field.
-
Income Type: 52% are Working, 24% are Commercial associates.
-
Education: 67% have Secondary / secondary special education, 28% have Higher education.
-
Marital Status: 68% are Married, 15% are Single / not married.
-
Housing Type: 89% live in a House / apartment.
-
Occupation: 32% of occupation data is missing, while Laborers account for 17% of specified occupations.
-
Family Members: The average family size is 2.16, with a maximum of 15 members.
-
Birthday_count has 1% missing entries.
-
Fields like Mobile_phone, Work_Phone, Phone, and EMAIL_ID indicate contact availability. All records show presence of a mobile phone.
-
Usage
This dataset is ideally suited for:
-
Developing and training binary classification models to predict credit card approval.
-
Performing Exploratory Data Analysis (EDA) to understand the relationships between various applicant features and approval outcomes.
-
Practising data cleaning and preprocessing techniques for real-world datasets.
-
Implementing and evaluating machine learning algorithms such as Random Forest.
-
Conducting model explainability studies to understand feature importance in credit decisions.
-
Educational purposes in data science and machine learning courses.
-
Coverage
The dataset provides demographic and financial information for a range of individuals.
-
Demographic Scope: Includes details such as gender, age (derived from Birthday_count), marital status, education level, family size, and occupation.
-
Financial Scope: Covers annual income, income type, and property/car ownership.
-
Geographic Scope: A specific geographic region is not stated for this dataset.
-
Time Range: A specific time range for data collection is not stated. Birthday and employment dates are given as backward counts from an unspecified current day. Data is available for a variety of groups, including those with different income levels, educational backgrounds, and family structures.
-
License
CC0: Public Domain
-
Who Can Use It
-
Data Scientists and Machine Learning Engineers: To build, train, and evaluate predictive models for credit risk assessment.
-
Students and Researchers: For academic projects focused on classification problems, data analysis, and model interpretation.
-
Financial Analysts: To gain insights into factors influencing credit approval and to support decision-making processes.
-
Anyone interested in practising data cleaning and preparation for a machine learning task.
-
Dataset Name Suggestions
-
Credit Application Approval Prediction Data
-
Client Credit Status Classification Dataset
-
Financial Applicant Assessment Data
-
Credit Card Decision Support Dataset
-
Personal Finance Classification Data
-
Attributes
Original Data Source: Client Credit Status Classification Dataset
QUALITY