{
  "id": 395864,
  "title": "Answers of questions are related ? - I checked",
  "url": "/competitions/predict-student-performance-from-game-play/discussion/395864",
  "author_name": "",
  "post_date": "2023-03-19T07:32:24.820711600Z",
  "votes": 6,
  "comment_count": 2,
  "views": 0,
  "content": "<ul>\n<li>Some part of my claim is maybe not correct . If you find mistakes, Please comment !!!</li>\n</ul>\n<p>In some discussions, I found a claim that \"the answer to question_x is useful in predicting the result of another question_y.\"</p>\n<p>So I calculated the Cohen's kappa score between the answers to question_x and question_y. <br>\nHere is my notebook↓<br>\n<a href=\"https://www.kaggle.com/code/whitelily/cohen-kappa-between-answers\" target=\"_blank\">https://www.kaggle.com/code/whitelily/cohen-kappa-between-answers</a></p>\n<p>As a result, <strong>no score was greater than 0.3. This (maybe) suggests a weak relationship between any answers to q_x and q_y.</strong></p>\n<p>Therefore, using the (predicted) answer may be useless in predicting.</p>\n<p>If you notice any problems with my thoughts, please do not hesitate to tell me!</p>",
  "messages": [
    {
      "id": "2187965",
      "postDate": "03/19/2023 07:32:24",
      "content": "<ul>\n<li>Some part of my claim is maybe not correct . If you find mistakes, Please comment !!!</li>\n</ul>\n<p>In some discussions, I found a claim that \"the answer to question_x is useful in predicting the result of another question_y.\"</p>\n<p>So I calculated the Cohen's kappa score between the answers to question_x and question_y. <br>\nHere is my notebook↓<br>\n<a href=\"https://www.kaggle.com/code/whitelily/cohen-kappa-between-answers\" target=\"_blank\">https://www.kaggle.com/code/whitelily/cohen-kappa-between-answers</a></p>\n<p>As a result, <strong>no score was greater than 0.3. This (maybe) suggests a weak relationship between any answers to q_x and q_y.</strong></p>\n<p>Therefore, using the (predicted) answer may be useless in predicting.</p>\n<p>If you notice any problems with my thoughts, please do not hesitate to tell me!</p>",
      "rawMarkdown": "* Some part of my claim is maybe not correct . If you find mistakes, Please comment !!!\n\nIn some discussions, I found a claim that \"the answer to question_x is useful in predicting the result of another question_y.\"\n\nSo I calculated the Cohen's kappa score between the answers to question_x and question_y. \nHere is my notebook↓\nhttps://www.kaggle.com/code/whitelily/cohen-kappa-between-answers\n\n\nAs a result, **no score was greater than 0.3. This (maybe) suggests a weak relationship between any answers to q_x and q_y.**\n\nTherefore, using the (predicted) answer may be useless in predicting.\n\nIf you notice any problems with my thoughts, please do not hesitate to tell me!",
      "votes": null
    },
    {
      "id": "2188558",
      "postDate": "03/19/2023 18:43:13",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/whitelily\" target=\"_blank\">@whitelily</a>, I tried to add the predictions to the questions from the previous groups to the features for the second and third group and it didn't help the score, so it goes in your direction. </p>",
      "rawMarkdown": "Hi @whitelily, I tried to add the predictions to the questions from the previous groups to the features for the second and third group and it didn't help the score, so it goes in your direction.",
      "votes": null
    },
    {
      "id": "2188600",
      "postDate": "03/19/2023 19:38:57",
      "content": "<p>It does help, only if you use it under the correct way. :)</p>",
      "rawMarkdown": "It does help, only if you use it under the correct way. :)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2188558,
      "author_name": "gehallak",
      "author_url": "",
      "post_date": "03/19/2023 18:43:13",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/whitelily\" target=\"_blank\">@whitelily</a>, I tried to add the predictions to the questions from the previous groups to the features for the second and third group and it didn't help the score, so it goes in your direction. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2188600,
      "author_name": "takanashihumbert",
      "author_url": "",
      "post_date": "03/19/2023 19:38:57",
      "content": "<p>It does help, only if you use it under the correct way. :)</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2187965": "* Some part of my claim is maybe not correct . If you find mistakes, Please comment !!!\n\nIn some discussions, I found a claim that \"the answer to question_x is useful in predicting the result of another question_y.\"\n\nSo I calculated the Cohen's kappa score between the answers to question_x and question_y. \nHere is my notebook↓\nhttps://www.kaggle.com/code/whitelily/cohen-kappa-between-answers\n\n\nAs a result, **no score was greater than 0.3. This (maybe) suggests a weak relationship between any answers to q_x and q_y.**\n\nTherefore, using the (predicted) answer may be useless in predicting.\n\nIf you notice any problems with my thoughts, please do not hesitate to tell me!",
    "2188558": "Hi @whitelily, I tried to add the predictions to the questions from the previous groups to the features for the second and third group and it didn't help the score, so it goes in your direction.",
    "2188600": "It does help, only if you use it under the correct way. :)"
  },
  "source": "meta"
}