{
  "id": 314035,
  "title": "Same Transaction with Different Prices or Channels",
  "url": "/competitions/h-and-m-personalized-fashion-recommendations/discussion/314035",
  "author_name": "",
  "post_date": "2022-03-20T14:54:24.848192200Z",
  "votes": 12,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hi everybody. I know there are some duplicate rows in <code>transactions_train.csv</code> indicating the <strong>multiple purchases</strong> of the same item (from <a href=\"https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/data\" target=\"_blank\">data page</a>).<br>\nHowever, I also find that the same transaction (<em>i.e.</em>, <code>(t_dat, customer_id, article_id)</code>) might have different <code>price</code> or different <code>sales_channel_id</code>. Following are my assumption:</p>\n<h4>1. Same Transaction with Different Prices</h4>\n<p>Example: <code>(2018-09-20, 003ac6eff3ca20cbe4e14a106a5e28b7551616005de8f9dd0f9455e5543b31ac, 0617322003)</code> has prices 0.033881 and 0.018627.<br>\nI think this might be because a <strong>discount</strong> is taken. Hence, the second item is cheaper.</p>\n<h4>2. Same Transaction with Different Sales Channels</h4>\n<p>Example: <code>(2018-09-20, 25952508d3a9fd13159404c54d7b9d1c656fdbdbb5d2e1f61b04fdb8c30432ac, 06650950002)</code> has channels 1 (store) and 2 (online).<br>\nMaybe, a customer buys the same item again online when he/she goes home and finds the article really suits. I can't come up with other idea now.</p>\n<p>I'm looking forward to your sharing and idea. If there's anything I miss or misunderstand, please let me know. Thanks a lot. </p>",
  "messages": [
    {
      "id": "1729821",
      "postDate": "03/20/2022 14:54:24",
      "content": "<p>Hi everybody. I know there are some duplicate rows in <code>transactions_train.csv</code> indicating the <strong>multiple purchases</strong> of the same item (from <a href=\"https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/data\" target=\"_blank\">data page</a>).<br>\nHowever, I also find that the same transaction (<em>i.e.</em>, <code>(t_dat, customer_id, article_id)</code>) might have different <code>price</code> or different <code>sales_channel_id</code>. Following are my assumption:</p>\n<h4>1. Same Transaction with Different Prices</h4>\n<p>Example: <code>(2018-09-20, 003ac6eff3ca20cbe4e14a106a5e28b7551616005de8f9dd0f9455e5543b31ac, 0617322003)</code> has prices 0.033881 and 0.018627.<br>\nI think this might be because a <strong>discount</strong> is taken. Hence, the second item is cheaper.</p>\n<h4>2. Same Transaction with Different Sales Channels</h4>\n<p>Example: <code>(2018-09-20, 25952508d3a9fd13159404c54d7b9d1c656fdbdbb5d2e1f61b04fdb8c30432ac, 06650950002)</code> has channels 1 (store) and 2 (online).<br>\nMaybe, a customer buys the same item again online when he/she goes home and finds the article really suits. I can't come up with other idea now.</p>\n<p>I'm looking forward to your sharing and idea. If there's anything I miss or misunderstand, please let me know. Thanks a lot. </p>",
      "rawMarkdown": "Hi everybody. I know there are some duplicate rows in `transactions_train.csv` indicating the **multiple purchases** of the same item (from [data page](https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/data)).\nHowever, I also find that the same transaction (*i.e.*, `(t_dat, customer_id, article_id)`) might have different `price` or different `sales_channel_id`. Following are my assumption:\n\n#### 1. Same Transaction with Different Prices\nExample: `(2018-09-20, 003ac6eff3ca20cbe4e14a106a5e28b7551616005de8f9dd0f9455e5543b31ac, 0617322003)` has prices 0.033881 and 0.018627.\nI think this might be because a **discount** is taken. Hence, the second item is cheaper.\n#### 2. Same Transaction with Different Sales Channels\nExample: `(2018-09-20, 25952508d3a9fd13159404c54d7b9d1c656fdbdbb5d2e1f61b04fdb8c30432ac, 06650950002)` has channels 1 (store) and 2 (online).\nMaybe, a customer buys the same item again online when he/she goes home and finds the article really suits. I can't come up with other idea now.\n\nI'm looking forward to your sharing and idea. If there's anything I miss or misunderstand, please let me know. Thanks a lot.",
      "votes": null
    },
    {
      "id": "1730123",
      "postDate": "03/21/2022 00:21:46",
      "content": "<p>The No.1 could be buy one, get another at 50% off. </p>\n<p>The No.2 could be a purchase for a family member or friend.</p>\n<p>Both these items are everyday Ladieswear.</p>",
      "rawMarkdown": "The No.1 could be buy one, get another at 50% off. \n\nThe No.2 could be a purchase for a family member or friend.\n\nBoth these items are everyday Ladieswear.",
      "votes": null
    },
    {
      "id": "1731780",
      "postDate": "03/22/2022 17:18:54",
      "content": "<p><a href=\"https://www.kaggle.com/atulverma\" target=\"_blank\">@atulverma</a>, thanks for your reply. It sounds reasonable.</p>",
      "rawMarkdown": "atulverma, thanks for your reply. It sounds reasonable.",
      "votes": null
    },
    {
      "id": "1735406",
      "postDate": "03/26/2022 07:59:05",
      "content": "<p>Isn't it the opposite: 1 for online &amp; 2 for offline  (as confirmed by the host in the Q&amp;A <a href=\"https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001\" target=\"_blank\">https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001</a> ) ? </p>",
      "rawMarkdown": "Isn't it the opposite: 1 for online & 2 for offline  (as confirmed by the host in the Q&A https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001 ) ?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1730123,
      "author_name": "atulverma",
      "author_url": "",
      "post_date": "03/21/2022 00:21:46",
      "content": "<p>The No.1 could be buy one, get another at 50% off. </p>\n<p>The No.2 could be a purchase for a family member or friend.</p>\n<p>Both these items are everyday Ladieswear.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1731780,
          "author_name": "abaojiang",
          "author_url": "",
          "post_date": "03/22/2022 17:18:54",
          "content": "<p><a href=\"https://www.kaggle.com/atulverma\" target=\"_blank\">@atulverma</a>, thanks for your reply. It sounds reasonable.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1735406,
      "author_name": "abelmebkhout",
      "author_url": "",
      "post_date": "03/26/2022 07:59:05",
      "content": "<p>Isn't it the opposite: 1 for online &amp; 2 for offline  (as confirmed by the host in the Q&amp;A <a href=\"https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001\" target=\"_blank\">https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001</a> ) ? </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1729821": "Hi everybody. I know there are some duplicate rows in `transactions_train.csv` indicating the **multiple purchases** of the same item (from [data page](https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/data)).\nHowever, I also find that the same transaction (*i.e.*, `(t_dat, customer_id, article_id)`) might have different `price` or different `sales_channel_id`. Following are my assumption:\n\n#### 1. Same Transaction with Different Prices\nExample: `(2018-09-20, 003ac6eff3ca20cbe4e14a106a5e28b7551616005de8f9dd0f9455e5543b31ac, 0617322003)` has prices 0.033881 and 0.018627.\nI think this might be because a **discount** is taken. Hence, the second item is cheaper.\n#### 2. Same Transaction with Different Sales Channels\nExample: `(2018-09-20, 25952508d3a9fd13159404c54d7b9d1c656fdbdbb5d2e1f61b04fdb8c30432ac, 06650950002)` has channels 1 (store) and 2 (online).\nMaybe, a customer buys the same item again online when he/she goes home and finds the article really suits. I can't come up with other idea now.\n\nI'm looking forward to your sharing and idea. If there's anything I miss or misunderstand, please let me know. Thanks a lot.",
    "1730123": "The No.1 could be buy one, get another at 50% off. \n\nThe No.2 could be a purchase for a family member or friend.\n\nBoth these items are everyday Ladieswear.",
    "1731780": "atulverma, thanks for your reply. It sounds reasonable.",
    "1735406": "Isn't it the opposite: 1 for online & 2 for offline  (as confirmed by the host in the Q&A https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations/discussion/307001 ) ?"
  },
  "source": "meta"
}