{
  "fetched_at": "2026-10-20T00:43:15Z",
  "source_url": "https://www.kaggle.com/competitions/churn-competition-pdc/data",
  "pages": [
    {
      "name": "Evaluation",
      "content": "Submissions are evaluated on [area under the ROC curve](http://en.wikipedia.org/wiki/Receiver_operating_characteristic) between the predicted probability and the observed target.\n\n**AUC-ROC** stands for *Area Under the Receiver Operating Characteristic Curve*. It is a performance measurement used for classification problems, especially when dealing with imbalanced datasets.\n\n**- ROC Curve:** The ROC curve is a graphical representation of a model's true positive rate (sensitivity) versus its false positive rate (1-specificity) across different threshold settings.\n\n**- AUC: **The Area Under the Curve (AUC) quantifies the overall ability of the model to discriminate between the positive class (churned) and the negative class (not churned).\n\nAn AUC of:\n\n- 0.5 indicates a model with no discrimination ability (random guessing).\n- 1.0 indicates a perfect model.\n\n>*In this competition, the higher the AUC-ROC score, the better the model is at correctly ranking customers by their likelihood of churning.*\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F13666066%2Fb19553b503d5ce53e46a98ae2ece3c6f%2FScreenshot%202024-08-25%20030523.jpg?generation=1724537152529540&alt=media)",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "# Competition Rules\n\n**1. Notebook Submissions Only:** Participants are required to submit their solutions through Kaggle Notebooks. Submissions from local files or external sources are not allowed. This ensures a consistent environment for all competitors.\n\n**2. No Internet Access:** Internet access is disabled in the competition notebooks to maintain a fair and controlled environment. Participants must rely solely on the provided dataset.\n\n**3. No Data Augmentation or External Data:** The use of data augmentation techniques or any additional external data is strictly prohibited. You must only use the dataset provided by the competition organizers.\n\n**4. One Account per Participant:** Each participant is allowed to compete using a single Kaggle account. Creating multiple accounts for the purpose of gaining an unfair advantage is strictly forbidden.\n\n**5. Private Sharing:** Private sharing of code or data outside of public Kaggle Kernels is prohibited. Collaborations should be conducted transparently and within the rules of Kaggle.\n\n**6. Submission Limits:** Participants are allowed a specific number of submissions per day (e.g., 25 submissions). The best submission of the day will count towards your final score.\n\n**7. Respect the Code of Conduct:** All participants are expected to maintain a high standard of behavior, respect fellow competitors, and adhere to the Kaggle Community Guidelines.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "### Data Description\n\nThe dataset provided for this competition contains information about customers of a financial institution, designed to help you build a model to predict customer churn. The data includes various attributes related to the customers' demographics, account information, and transaction history. Here's a brief overview of the key columns in the dataset:\n\n- **CustomerId**: A unique identifier for each customer.\n- **Surname**: The last name of the customer (irrelevant for modeling but useful for identification).\n- **CreditScore**: A numerical score representing the customer's creditworthiness.\n- **Geography**: The country of residence of the customer (e.g., France, Germany, Spain).\n- **Gender**: The gender of the customer (Male or Female).\n- **Age**: The age of the customer.\n- **Tenure**: The number of years the customer has been with the bank.\n- **Balance**: The current balance in the customer's account.\n- **NumOfProducts**: The number of products the customer has with the bank.\n- **HasCrCard**: Whether the customer has a credit card (1 = Yes, 0 = No).\n- **IsActiveMember**: Whether the customer is an active member (1 = Yes, 0 = No).\n- **EstimatedSalary**: The estimated annual salary of the customer.\n- **Exited**: The target variable, indicating whether the customer has churned (1 = Yes, 0 = No).\n\n### Purpose of the Data\n\nThe data is structured to help you identify patterns that might indicate why a customer would decide to leave the bank. By analyzing these attributes, your task is to build a model that predicts the likelihood of customer churn based on the features provided.\n\n### Training and Test Sets\n\n- **train.csv**: Contains the labeled data, including the `Exited` column, which indicates whether the customer churned. This is the data you will use to train your model.\n- **test.csv**: Contains the same features as the training set but without the `Exited` column. Your task is to predict the probability of churn for each customer in this dataset.\n\nThis dataset provides a rich set of features that, when analyzed and modeled effectively, can yield valuable insights into customer behavior and churn prediction.\n\n\n## Files\n\n*   **train.csv** - the training dataset; `Exited` is the binary target\n*   **test.csv** - the test dataset; your objective is to predict the probability of `Exited`\n*   **sample_submission.csv** - a sample submission file in the correct format",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "**Welcome to our first Kaggle competition!**\n\nIn this Kaggle competition, participants are tasked with building predictive models to identify customers who are likely to churn. Churn refers to the event when a customer decides to leave a service or cancel a subscription. Predicting customer churn is crucial for businesses, as retaining existing customers is often more cost-effective than acquiring new ones.\n\nYou are provided with a dataset containing various customer attributes, such as demographic information, account details, and transactional behaviors. The goal is to develop a machine learning model that can accurately predict whether a customer will churn based on these attributes.\n\n# **Why is Customer Churn Prediction Important?**\n\nCustomer churn is a significant concern for businesses across various industries, from telecommunications to banking and subscription services. The loss of a customer not only results in immediate revenue loss but also impacts long-term profitability. Moreover, acquiring a new customer can be 5 to 25 times more expensive than retaining an existing one. Therefore, understanding the factors that lead to customer churn and being able to predict it in advance can provide businesses with critical insights to implement targeted retention strategies.\n\n## **Learning Outcomes:**\n\nThis competition offers a unique opportunity for students to apply their data science and machine learning skills to a real-world problem with tangible business implications. By participating in this challenge, students will:\n\n- **Develop Predictive Models:** Gain hands-on experience in building and evaluating machine learning models.\n- **Feature Engineering:** Learn how to engineer meaningful features that can significantly enhance model performance.\n- **Model Evaluation and Optimization:** Explore different algorithms and techniques to optimize the model and achieve the best performance.\n- **Critical Thinking:** Understand the business context of customer churn and think critically about how predictive analytics can drive business decisions.\n\n## **Challenge Yourself:**\n\nWhether you are a beginner or an experienced data scientist, this competition will challenge you to think creatively and push the boundaries of your problem-solving skills. It’s an excellent opportunity to showcase your abilities, learn from others in the community, and contribute to solving a problem that is central to business success.\n\nGood luck, and may the best model win!\n\n### Happy Learning...!!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}