{
  "fetched_at": "2026-10-19T23:23:34Z",
  "source_url": "https://www.kaggle.com/competitions/clevernlp/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The data was a list of 41,087 questions, some of which were suggested by users, and some (about 10%) were created by editors. The challenge was to distinguish questions prepared by professional authors from questions suggested by game users.\n\nFiles:\n- train.csv contains id of question and answer \n- data.csv containt id of question and question on Russian\n- test.csc contains only questions and you shoul upload your solution for this dataset in kaggle",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "The data was a list of 41,087 questions, some of which were suggested by users, and some (about 10%) were created by editors. The challenge is distinguish questions prepared by professional authors from questions suggested by game users.\n\n\"Clever\" is an entertaining game with prizes that took place from March 14 to December 31, 2018 in a special mobile application. In the game, the participants had to choose the correct answer to the question asked from the three proposed options. Players who answered correctly all 12-15 questions of the game session equally divided the money fund being played.\n\nThe data for this problem was taken from the ML Boot Camp platform, from the materials of the VK Cup competition.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Evaluation metric: ROC-AUC.\n\nAUC - ROC curve is a performance measurement for the classification problems at various threshold settings. ROC is a probability curve and AUC represents the degree or measure of separability. It tells how much the model is capable of distinguishing between classes. Higher the AUC, the better the model is at predicting 0 classes as 0 and 1 classes as 1. By analogy, the Higher the AUC, the better the model is at distinguishing between patients with the disease and no disease.\n\n![ROC-AUC](https://miro.medium.com/max/451/1*pk05QGzoWhCgRiiFbz-oKQ.png)\n\nThe ROC curve is plotted with TPR against the FPR where TPR is on the y-axis and FPR is on the x-axis.\n\n**Confusion matrix:**\n```markdown\nTrue positive (TP) |     False positive (FP)\n--------------------------------------------\nFalse negative (FN) |     True negative (TN)\n```\n---\n\n**TPR = TP/(TP+FN)**\n\n**FRR = FP/(TN+FP)**\n\n`from sklearn.metrics import roc_auc_score`\n\n## Submission Format\n\nSubmission Format is csv.\n\nThe file should contain a header and have the following format:\n```\nAuthorId,DuplicateAuthorIds\n1,1\n```",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n- Learn!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}