{
  "id": 313942,
  "title": "Product discounts",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/313942",
  "author_name": "",
  "post_date": "2022-03-19T19:43:23.714913300Z",
  "votes": 7,
  "comment_count": 3,
  "views": 0,
  "content": "<p>In the train dataset there are different prices for the same product. I guess the price varies in different locations etc, but also with discounts. In the prediction it might be helpful to have the current products prices.  </p>",
  "messages": [
    {
      "id": "1729230",
      "postDate": "03/19/2022 19:43:23",
      "content": "<p>In the train dataset there are different prices for the same product. I guess the price varies in different locations etc, but also with discounts. In the prediction it might be helpful to have the current products prices.  </p>",
      "rawMarkdown": "In the train dataset there are different prices for the same product. I guess the price varies in different locations etc, but also with discounts. In the prediction it might be helpful to have the current products prices.",
      "votes": null
    },
    {
      "id": "1731135",
      "postDate": "03/22/2022 02:37:54",
      "content": "<p>Good idea. If you compute the mean price per item (or linear regression the current price from historic trend). Then we can check whether an item is currently on sale or not (by comparing to what it should be). This will probably affect how many customers are currently buying the item.</p>",
      "rawMarkdown": "Good idea. If you compute the mean price per item (or linear regression the current price from historic trend). Then we can check whether an item is currently on sale or not (by comparing to what it should be). This will probably affect how many customers are currently buying the item.",
      "votes": null
    },
    {
      "id": "1731190",
      "postDate": "03/22/2022 04:45:16",
      "content": "<p>But you wouldn't know whether the article will be on sale or how much sale during target week. I believe, <a href=\"https://www.kaggle.com/eldadp100\" target=\"_blank\">@eldadp100</a> , is asking for that.</p>",
      "rawMarkdown": "But you wouldn't know whether the article will be on sale or how much sale during target week. I believe, @eldadp100 , is asking for that.",
      "votes": null
    },
    {
      "id": "1731193",
      "postDate": "03/22/2022 04:54:02",
      "content": "<p>Yes and no. If a product is on sale for the last day (or few) of train data, it may still be on sale in the test data.</p>\n<p>We can also try to learn the pattern of sales from train data. I do not believe the host will provide more features for test data. So we must do the best we can with what we have.</p>",
      "rawMarkdown": "Yes and no. If a product is on sale for the last day (or few) of train data, it may still be on sale in the test data.\n\nWe can also try to learn the pattern of sales from train data. I do not believe the host will provide more features for test data. So we must do the best we can with what we have.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1731135,
      "author_name": "cdeotte",
      "author_url": "",
      "post_date": "03/22/2022 02:37:54",
      "content": "<p>Good idea. If you compute the mean price per item (or linear regression the current price from historic trend). Then we can check whether an item is currently on sale or not (by comparing to what it should be). This will probably affect how many customers are currently buying the item.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1731190,
          "author_name": "atulverma",
          "author_url": "",
          "post_date": "03/22/2022 04:45:16",
          "content": "<p>But you wouldn't know whether the article will be on sale or how much sale during target week. I believe, <a href=\"https://www.kaggle.com/eldadp100\" target=\"_blank\">@eldadp100</a> , is asking for that.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1731193,
          "author_name": "cdeotte",
          "author_url": "",
          "post_date": "03/22/2022 04:54:02",
          "content": "<p>Yes and no. If a product is on sale for the last day (or few) of train data, it may still be on sale in the test data.</p>\n<p>We can also try to learn the pattern of sales from train data. I do not believe the host will provide more features for test data. So we must do the best we can with what we have.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1729230": "In the train dataset there are different prices for the same product. I guess the price varies in different locations etc, but also with discounts. In the prediction it might be helpful to have the current products prices.",
    "1731135": "Good idea. If you compute the mean price per item (or linear regression the current price from historic trend). Then we can check whether an item is currently on sale or not (by comparing to what it should be). This will probably affect how many customers are currently buying the item.",
    "1731190": "But you wouldn't know whether the article will be on sale or how much sale during target week. I believe, @eldadp100 , is asking for that.",
    "1731193": "Yes and no. If a product is on sale for the last day (or few) of train data, it may still be on sale in the test data.\n\nWe can also try to learn the pattern of sales from train data. I do not believe the host will provide more features for test data. So we must do the best we can with what we have."
  },
  "source": "meta"
}