{
  "fetched_at": "2026-10-20T05:16:12Z",
  "source_url": "https://www.kaggle.com/competitions/bads-dab-regression/data",
  "pages": [
    {
      "name": "data-description",
      "content": "This dataset comprises loans issued via the **Lending Club platform**, an individual-to-individual lending platform. However, it's crucial to acknowledge that not all loans carry the same level of risk. Individuals with a strong likelihood of repaying their loans are more likely to secure a loan with a lower interest rate, unlike those perceived as higher risk. Consequently, individuals categorized as highly risky may either not receive a loan offer at all or decline the offer due to the associated high interest rate. It is essential to remember that this dataset solely represents loans that have been actually disbursed, thus it should not be mistaken for loan applications\n\nThe problem at hand is to predict the loan interest rate. \n\n## Files\n\n*   **train.csv** - the training set\n*   **test.csv** - the test set\n*   **sample_submission.csv** - a sample submission file in the correct format\n\n## Columns\n- `loan_number` Unique identifier\n- `emp_length `\t Number of years in the job, rounded down. If longer than 10 years, then this is represented by the value 10.\t\t\n- `home_ownership `\t The ownership status of the applicant's residence.\t\t\n- `annual_income` \t Annual income.\t\t\n- `verified_income` \t Type of verification of the applicant's income.\t\t\n- `debt_to_income` \t Debt\tto\tincome ratio.\n- `delinq_2y` \t Delinquencies on lines of credit in the last 2 years.\t\t\n- `total_credit_limit` \t Total available credit, e.g. if only credit cards, then the total of all the credit limits. This excludes a mortgage.\t\t\n- `total_credit_utilized` \t Total credit balance, excluding a mortgage.\t\t\n- `public_record_bankrupt` \t Number of bankruptcies listed in the public record for this applicant.\t\t\n- `loan_purpose` \t The category for the purpose of the loan.\t\t\n- `application_type` \t The type of application: either individual or joint.\t\t\n- `loan_amount` \t The amount of the loan the applicant received.\t\t\n- `term` \t The number of months of the loan the applicant received.\t\t\n- `issue_month` \t Month the loan was issued.\t\t\n- `interest_rate `\t **Target**  Interest rate of the loan the applicant received.\t\t",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "# In-Class Contest - Interest Rate Prediction\n\n\n\n## Contest Objective\nThe objective of the contest is to develop predictive models that accurately estimate the interest rate based on various input variables. Participants are expected to create models that generalize well to unseen data and minimize the RMSE between the predicted interest rates and the actual interest rates.\n\n## Evaluation Metric: RMSE\nRMSE is a commonly used evaluation metric for regression tasks. It measures the average deviation between the predicted values and the actual values, taking the square root of the mean of the squared differences. In the context of this contest, the RMSE will be calculated to assess the accuracy of the interest rate predictions.\n\n## Submission Format\nParticipants are required to submit a CSV file containing their predictions for the interest rates corresponding to the test dataset provided by Kaggle. The submission file should include a unique identifier for each test instance along with the predicted interest rate.\n\nThe submission format should follow the guidelines specified in the contest instructions. Failure to adhere to the format may result in disqualification.\n\n## Evaluation Process\nOnce the participants submit their predictions, the Kaggle platform will calculate the RMSE by comparing the predicted interest rates to the actual interest rates for a separate test dataset, which is not available to the participants during the contest.\n\nThe submissions will be ranked based on the RMSE scores, with lower scores indicating better predictions. The leaderboard will display the rankings and scores of the participants throughout the contest duration.\n\n## Final Evaluation and Winner Selection\nAt the end of the contest period, the final evaluation will be conducted based on the performance of the participants' models on a holdout dataset, which is kept hidden until the contest ends. The participants with the lowest RMSE scores on this holdout dataset will be considered the winners of the Kaggle contest.\n\n## Conclusion\nThe Kaggle contest on interest rate prediction involves developing accurate models to estimate interest rates based on provided input variables. The evaluation metric, RMSE, will be used to assess the accuracy of the predictions. Participants are required to submit their predictions in the specified format, and the final winners will be determined based on their performance on the holdout dataset.\n\nGood luck to all the participants!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "RMSE is a commonly used evaluation metric for regression tasks. It measures the average deviation between the predicted values and the actual values, taking the square root of the mean of the squared differences. In the context of this contest, the RMSE will be calculated to assess the accuracy of the interest rate predictions.\n\nRMSE is defined as the square root of the mean squared error (MSE), which is the average squared difference between the predicted values of the model and the true values of the data.\n\nThe RMSE is calculated using the following formula:\n\n$$\\displaystyle RMSE = \\sqrt{\\frac{\\Sigma_{i=1}^{i=n}(y_i-\\hat{y_i})}{n}}$$\n\nWhere n is the number of observations in the data, y_i is the true value of the i-th observation, and ŷ_i is the predicted value of the i-th observation.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": " - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}