{
  "fetched_at": "2026-10-19T16:11:47Z",
  "source_url": "https://www.kaggle.com/competitions/dauphine-datascience-ecommerce-2021-1/data",
  "pages": [
    {
      "name": "data-description",
      "content": "# [train|test]\n\nOne row = one product on one given day\n\n - `s_date`, `s_product` : primary key\n - `s_brand`: brand of the product\n - `s_family`: highest level category  on the website (Gardening, Tools, ...)\n - `s_lowest_category`: lowest level category on the website (Drills, swimming pools, ...)\n - `s_provider`: seller of the product\n - `s_retail_price`: observed retail price\n - `s_shipping_time`: estimated shipping time\n - `s_weight`: weight of the product\n - `m_nb_ratings`: number of ratings of the product\n - `m_avg_rating`: average rating of the product\n - `m_has_pdf`: product has a pdf documentation\n - `m_length_description`: length of the product description\n - `m_unit_price_vat_inc`: product price including VAT\n - `m_nb_images`: number of images of the product on the website\n - `m_total_vu`: number of views of the product page (not present in test set)\n - `conversion_rate` : `m_total_transaction / m_total_vu` (our target not present in test set)\n\nNote that you can access the number of orders by multiplying `m_total_vu` and `conversion_rate`\n\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h2>Predict conversion rate</h2>\n\nThe goal of the competition is to predict the **conversion rate** of a product on a given day.\n\nThe training dataset contains data on all products of our top 10 categories for January 2020.\n\nThe test dataset contains data for February 2020 (until Feb 11).",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "**Weighted Root Mean Squared Error (WRMSE)**\n<p>Evaluation will be performed using Weighted Root Mean Square Error:</p>\n<p>\\[ \\epsilon = \\sqrt{\\frac{\\sum_{i=1}^n v_i (y_i - \\hat{y}_i)^2 }{\\sum_{i=1}^n v_i} } \\]</p>\n<p>where v is the number of views (m_total_vu) of the product page, y is the conversion rate of the product and y hat its predicted value.</p>\n\n\n**Submission File**\nFor every (s\\_date, s\\_product) in the dataset, submission files should contain two columns: \nid and conversion_rate\n\nThe file should contain a header and have the following format:\n\nid,conversion_rate\n0,0.03\n1,0.03\n2,0.03\netc...",
      "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 - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}