{
  "id": 331172,
  "title": "Papers: Machine Learning + Credit default prediction",
  "url": "/competitions/amex-default-prediction/discussion/331172",
  "author_name": "",
  "post_date": "2022-06-16T03:47:39.166868900Z",
  "votes": 10,
  "comment_count": 1,
  "views": 0,
  "content": "<h3>Papers: Machine Learning + Credit default detection</h3>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/253a297504c47ab986d6f6147392858eab8cf0da\" target=\"_blank\">A Classification Approach of Neural Networks for Credit Card Default Detection\n</a></strong></p>\n<p>This article mainly tried to determine the factors that strongly predict the future default probability with neural network comparing with linear model shows the advantage of deep learning in financial area. </p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Artificial-neural-network-technique-for-improving-A-Ebiaredoh-Mienye-Esenogho/4d58460854dddf3f9cbc9c72c6df040c22ac9bcd\" target=\"_blank\">Artificial neural network technique for improving prediction of credit card default: A stacked sparse autoencoder approach</a></strong></p>\n<p>An unsupervised feature learning method to improve the performance of various classifiers using a stacked sparse autoencoder (SSAE) was proposed and showed superior performance over other methods. </p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Comparison-of-Different-Ensemble-Methods-in-Credit-Faraj-Mahmud/1c941db8630d2363143250c9fddfbb739e05f23f\" target=\"_blank\">Comparison of Different Ensemble Methods in Credit Card Default Prediction</a></strong></p>\n<p>The results of this study show that ensemble methods consistently outperform Neural Networks and other machine learning algorithms in terms of F1 score and area under receiver operating characteristic curve regardless of balancing the dataset or ignoring the imbalance.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Forecasting-of-Credit-Card-Default-Based-on-Random-Gao-Wen/b1c7b36e4c7d0d3232e36a325929c7c3ce8fc110\" target=\"_blank\">Forecasting of Credit Card Default Based on Incremental Random Forest</a></strong></p>\n<p>The incremental random forest algorithm is proposed for the classification and prediction problem of dynamically increasing data and has relatively better performance in the experiment of predicting the default behavior ofcredit card customers based on a batch of credit card holder data of a bank in Taiwan.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/The-Application-of-Machine-Learning-Algorithms-in-Yu/4e80f5e510b016c2a492e4c6324ff6ef51aa5f69\" target=\"_blank\">The Application of Machine Learning Algorithms in Credit Card Default Prediction</a></strong></p>\n<p>Several classical machine learning algorithms are employed, including logistic regression, decision tree, decision tree and ensemble learning, to build credit default prediction models to solve the problem of unbalanced data and the results show that random forest models with weight is the best, which achieved an accuracy of 82.12%.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Application-of-Machine-Learning-Algorithms-in-Card-Husejinovi%C4%87-Ke%C4%8Do/56d31fbacdf19f2221206f3d519effa9fbc72385\" target=\"_blank\">Application of Machine Learning Algorithms in Credit Card Default Payment Prediction</a></strong></p>\n<p>The performance of machine learning methods on credit card default payment prediction using logistic regression, C4.5 decision tree, support vector machines, naive Bayes, k-nearest neighbors algorithms, and ensemble learning methods voting, bagging and boosting is evaluated.</p>",
  "messages": [
    {
      "id": "1822067",
      "postDate": "06/16/2022 03:47:39",
      "content": "<h3>Papers: Machine Learning + Credit default detection</h3>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/253a297504c47ab986d6f6147392858eab8cf0da\" target=\"_blank\">A Classification Approach of Neural Networks for Credit Card Default Detection\n</a></strong></p>\n<p>This article mainly tried to determine the factors that strongly predict the future default probability with neural network comparing with linear model shows the advantage of deep learning in financial area. </p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Artificial-neural-network-technique-for-improving-A-Ebiaredoh-Mienye-Esenogho/4d58460854dddf3f9cbc9c72c6df040c22ac9bcd\" target=\"_blank\">Artificial neural network technique for improving prediction of credit card default: A stacked sparse autoencoder approach</a></strong></p>\n<p>An unsupervised feature learning method to improve the performance of various classifiers using a stacked sparse autoencoder (SSAE) was proposed and showed superior performance over other methods. </p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Comparison-of-Different-Ensemble-Methods-in-Credit-Faraj-Mahmud/1c941db8630d2363143250c9fddfbb739e05f23f\" target=\"_blank\">Comparison of Different Ensemble Methods in Credit Card Default Prediction</a></strong></p>\n<p>The results of this study show that ensemble methods consistently outperform Neural Networks and other machine learning algorithms in terms of F1 score and area under receiver operating characteristic curve regardless of balancing the dataset or ignoring the imbalance.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Forecasting-of-Credit-Card-Default-Based-on-Random-Gao-Wen/b1c7b36e4c7d0d3232e36a325929c7c3ce8fc110\" target=\"_blank\">Forecasting of Credit Card Default Based on Incremental Random Forest</a></strong></p>\n<p>The incremental random forest algorithm is proposed for the classification and prediction problem of dynamically increasing data and has relatively better performance in the experiment of predicting the default behavior ofcredit card customers based on a batch of credit card holder data of a bank in Taiwan.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/The-Application-of-Machine-Learning-Algorithms-in-Yu/4e80f5e510b016c2a492e4c6324ff6ef51aa5f69\" target=\"_blank\">The Application of Machine Learning Algorithms in Credit Card Default Prediction</a></strong></p>\n<p>Several classical machine learning algorithms are employed, including logistic regression, decision tree, decision tree and ensemble learning, to build credit default prediction models to solve the problem of unbalanced data and the results show that random forest models with weight is the best, which achieved an accuracy of 82.12%.</p>\n<p><strong><a href=\"https://www.semanticscholar.org/paper/Application-of-Machine-Learning-Algorithms-in-Card-Husejinovi%C4%87-Ke%C4%8Do/56d31fbacdf19f2221206f3d519effa9fbc72385\" target=\"_blank\">Application of Machine Learning Algorithms in Credit Card Default Payment Prediction</a></strong></p>\n<p>The performance of machine learning methods on credit card default payment prediction using logistic regression, C4.5 decision tree, support vector machines, naive Bayes, k-nearest neighbors algorithms, and ensemble learning methods voting, bagging and boosting is evaluated.</p>",
      "rawMarkdown": "### Papers: Machine Learning + Credit default detection\n\n\n**[A Classification Approach of Neural Networks for Credit Card Default Detection\n](https://www.semanticscholar.org/paper/253a297504c47ab986d6f6147392858eab8cf0da)**\n\nThis article mainly tried to determine the factors that strongly predict the future default probability with neural network comparing with linear model shows the advantage of deep learning in financial area. \n\n**[Artificial neural network technique for improving prediction of credit card default: A stacked sparse autoencoder approach](https://www.semanticscholar.org/paper/Artificial-neural-network-technique-for-improving-A-Ebiaredoh-Mienye-Esenogho/4d58460854dddf3f9cbc9c72c6df040c22ac9bcd)**\n\nAn unsupervised feature learning method to improve the performance of various classifiers using a stacked sparse autoencoder (SSAE) was proposed and showed superior performance over other methods. \n\n**[Comparison of Different Ensemble Methods in Credit Card Default Prediction](https://www.semanticscholar.org/paper/Comparison-of-Different-Ensemble-Methods-in-Credit-Faraj-Mahmud/1c941db8630d2363143250c9fddfbb739e05f23f)**\n\nThe results of this study show that ensemble methods consistently outperform Neural Networks and other machine learning algorithms in terms of F1 score and area under receiver operating characteristic curve regardless of balancing the dataset or ignoring the imbalance.\n\n**[Forecasting of Credit Card Default Based on Incremental Random Forest](https://www.semanticscholar.org/paper/Forecasting-of-Credit-Card-Default-Based-on-Random-Gao-Wen/b1c7b36e4c7d0d3232e36a325929c7c3ce8fc110)**\n\nThe incremental random forest algorithm is proposed for the classification and prediction problem of dynamically increasing data and has relatively better performance in the experiment of predicting the default behavior ofcredit card customers based on a batch of credit card holder data of a bank in Taiwan.\n\n**[The Application of Machine Learning Algorithms in Credit Card Default Prediction](https://www.semanticscholar.org/paper/The-Application-of-Machine-Learning-Algorithms-in-Yu/4e80f5e510b016c2a492e4c6324ff6ef51aa5f69)**\n\nSeveral classical machine learning algorithms are employed, including logistic regression, decision tree, decision tree and ensemble learning, to build credit default prediction models to solve the problem of unbalanced data and the results show that random forest models with weight is the best, which achieved an accuracy of 82.12%.\n\n**[Application of Machine Learning Algorithms in Credit Card Default Payment Prediction](https://www.semanticscholar.org/paper/Application-of-Machine-Learning-Algorithms-in-Card-Husejinovi%C4%87-Ke%C4%8Do/56d31fbacdf19f2221206f3d519effa9fbc72385)**\n\nThe performance of machine learning methods on credit card default payment prediction using logistic regression, C4.5 decision tree, support vector machines, naive Bayes, k-nearest neighbors algorithms, and ensemble learning methods voting, bagging and boosting is evaluated.",
      "votes": null
    },
    {
      "id": "1822082",
      "postDate": "06/16/2022 04:10:19",
      "content": "<p>For credit default prediction I'd also recommend Bart Baesens books</p>",
      "rawMarkdown": "For credit default prediction I'd also recommend Bart Baesens books",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1822082,
      "author_name": "denisplaj",
      "author_url": "",
      "post_date": "06/16/2022 04:10:19",
      "content": "<p>For credit default prediction I'd also recommend Bart Baesens books</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1822067": "### Papers: Machine Learning + Credit default detection\n\n\n**[A Classification Approach of Neural Networks for Credit Card Default Detection\n](https://www.semanticscholar.org/paper/253a297504c47ab986d6f6147392858eab8cf0da)**\n\nThis article mainly tried to determine the factors that strongly predict the future default probability with neural network comparing with linear model shows the advantage of deep learning in financial area. \n\n**[Artificial neural network technique for improving prediction of credit card default: A stacked sparse autoencoder approach](https://www.semanticscholar.org/paper/Artificial-neural-network-technique-for-improving-A-Ebiaredoh-Mienye-Esenogho/4d58460854dddf3f9cbc9c72c6df040c22ac9bcd)**\n\nAn unsupervised feature learning method to improve the performance of various classifiers using a stacked sparse autoencoder (SSAE) was proposed and showed superior performance over other methods. \n\n**[Comparison of Different Ensemble Methods in Credit Card Default Prediction](https://www.semanticscholar.org/paper/Comparison-of-Different-Ensemble-Methods-in-Credit-Faraj-Mahmud/1c941db8630d2363143250c9fddfbb739e05f23f)**\n\nThe results of this study show that ensemble methods consistently outperform Neural Networks and other machine learning algorithms in terms of F1 score and area under receiver operating characteristic curve regardless of balancing the dataset or ignoring the imbalance.\n\n**[Forecasting of Credit Card Default Based on Incremental Random Forest](https://www.semanticscholar.org/paper/Forecasting-of-Credit-Card-Default-Based-on-Random-Gao-Wen/b1c7b36e4c7d0d3232e36a325929c7c3ce8fc110)**\n\nThe incremental random forest algorithm is proposed for the classification and prediction problem of dynamically increasing data and has relatively better performance in the experiment of predicting the default behavior ofcredit card customers based on a batch of credit card holder data of a bank in Taiwan.\n\n**[The Application of Machine Learning Algorithms in Credit Card Default Prediction](https://www.semanticscholar.org/paper/The-Application-of-Machine-Learning-Algorithms-in-Yu/4e80f5e510b016c2a492e4c6324ff6ef51aa5f69)**\n\nSeveral classical machine learning algorithms are employed, including logistic regression, decision tree, decision tree and ensemble learning, to build credit default prediction models to solve the problem of unbalanced data and the results show that random forest models with weight is the best, which achieved an accuracy of 82.12%.\n\n**[Application of Machine Learning Algorithms in Credit Card Default Payment Prediction](https://www.semanticscholar.org/paper/Application-of-Machine-Learning-Algorithms-in-Card-Husejinovi%C4%87-Ke%C4%8Do/56d31fbacdf19f2221206f3d519effa9fbc72385)**\n\nThe performance of machine learning methods on credit card default payment prediction using logistic regression, C4.5 decision tree, support vector machines, naive Bayes, k-nearest neighbors algorithms, and ensemble learning methods voting, bagging and boosting is evaluated.",
    "1822082": "For credit default prediction I'd also recommend Bart Baesens books"
  },
  "source": "meta"
}