{
  "id": 205196,
  "title": "Cannot understand why `prior_explanations_avg` of current question is important",
  "url": "/competitions/riiid-test-answer-prediction/discussion/205196",
  "author_name": "william.wu",
  "post_date": "2020-12-19T00:38:31.970000",
  "votes": 1,
  "comment_count": 0,
  "views": 0,
  "content": "<p>I saw many public LGB notebooks using the features like <code>explanation_q_avg</code> and <code>elapsed_time_q_avg</code> in <a href=\"https://www.kaggle.com/ragnar123/riiid-model-lgbm\" target=\"_blank\">https://www.kaggle.com/ragnar123/riiid-model-lgbm</a> they are calculated based on the <code>prior_question_had_explanation</code> and <code>prior_question_elapsed_time</code> under each question group respectively. Because they question-related features, do they tell some information about the current question?</p>",
  "messages": [
    {
      "id": 1118330,
      "postDate": "2020-12-19T00:38:31.970Z",
      "content": "<p>I saw many public LGB notebooks using the features like <code>explanation_q_avg</code> and <code>elapsed_time_q_avg</code> in <a href=\"https://www.kaggle.com/ragnar123/riiid-model-lgbm\" target=\"_blank\">https://www.kaggle.com/ragnar123/riiid-model-lgbm</a> they are calculated based on the <code>prior_question_had_explanation</code> and <code>prior_question_elapsed_time</code> under each question group respectively. Because they question-related features, do they tell some information about the current question?</p>",
      "rawMarkdown": "I saw many public LGB notebooks using the features like `explanation_q_avg ` and `elapsed_time_q_avg ` in https://www.kaggle.com/ragnar123/riiid-model-lgbm they are calculated based on the `prior_question_had_explanation ` and `prior_question_elapsed_time ` under each question group respectively. Because they question-related features, do they tell some information about the current question?",
      "votes": 1
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1118330": "I saw many public LGB notebooks using the features like `explanation_q_avg ` and `elapsed_time_q_avg ` in https://www.kaggle.com/ragnar123/riiid-model-lgbm they are calculated based on the `prior_question_had_explanation ` and `prior_question_elapsed_time ` under each question group respectively. Because they question-related features, do they tell some information about the current question?"
  }
}