{
  "fetched_at": "2026-10-19T13:38:05Z",
  "source_url": "https://www.kaggle.com/competitions/csc-575-final-project-kaggle-winter-2019/data",
  "pages": [
    {
      "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
    },
    {
      "name": "data-description",
      "content": "<p>This data set contains a number of products and real customer search terms from Home Depot's website. The challenge is to predict a relevance score for the provided combinations of search terms and products. To create the ground truth labels, Home Depot has crowdsourced the search/product pairs to multiple human raters.</p>\n\n<p>The relevance is a number between 1 (not relevant) to 3 (highly relevant). For example, a search for \"AA battery\" would be considered highly relevant to a pack of size AA batteries (relevance = 3), mildly relevant to a cordless drill battery (relevance = 2), and not relevant to a snow shovel (relevance = 1).</p>\n\n<p>Each pair was evaluated by at least three human raters. The provided relevance scores are the average value of the ratings. There are three additional things to know about the ratings:\n<ul>\n<li>The specific instructions given to the raters is provided in \"relevance_instructions.docx\".\n<li>Raters did not have access to the attributes.\n<li>Raters had access to product images, while the competition does not include images.\n</ul>\nYour task is to predict the relevance for each pair listed in the test set. Note that the test set contains both seen and unseen search terms.\n</p>\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>train_new.csv</strong> - the training set, contains products, searches, and relevance scores\n<li><strong>test_new.csv</strong> - the test set, contains products and searches. You must predict the relevance for these pairs.\n<li><strong>product_descriptions_new.csv</strong> - contains a text description of each product. You may join this table to the training or test set via the product_uid.\n<li><strong>attributes_new.csv</strong> -  provides extended information about a subset of the products (typically representing detailed technical specifications). Not every product will have attributes.\n<li><strong>sample_submission_new.csv</strong> - a file showing the correct submission format\n<li><strong>relevance_instructions.docx</strong> - the instructions provided to human raters\n</ul>\n\n\n<h2>Data fields</h2>\n<ul>\n<li><strong>id</strong> - a unique Id field which represents a (search_term, product_uid) pair\n<li><strong>product_uid</strong> - an id for the products\n<li><strong>product_title</strong> - the product title\n<li><strong>product_description</strong> - the text description of the product (may contain HTML content)\n<li><strong>search_term</strong> - the search query\n<li><strong>relevance</strong> - the average of the relevance ratings for a given id\n<li><strong>name</strong> - an attribute name\n<li><strong>value</strong> - the attribute's value\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p>The task of this competition is to predict the relevance scores of product search results on <a href=\"https://www.homedepot.com/c/homepageregional1?s_tnt=114337:1:0\">homedepot.com</a>.  Specifically, you predict a relevance score for a given product_title-search_term pair.  For example, </p><pre>Product title: “Plastec 11 in. x 24 in. Rose Garden Wall Decor Steel”\nSearch term: “zen garden  decor”\nRelevance: 3\n</pre>\nThe data set contains a number of products (with titles, which are short descriptions) and real customer search queries from Home Depot's website.  There is another file that contains long descriptions of the products.  The challenge is to predict a relevance score for the provided combinations of search terms and products.  Home Depot had crowdsourced the search/product pairs to multiple human raters to create the ground truth labels.\n<p></p>\n<p>The competition closes on 3/17 (Sun) 6:59 pm Chicago time (which is 11:59 pm UTC/GMT).</p>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>The evaluation metric for this competition is Root Mean Squared Error (RMSE), which is defined as:\n\n</p><p>\\[ RMSE = \\sqrt{\\frac{1}{n} \\sum_{i=1}^{n} (y_{i} - \\hat{y_{i}})^{2} }\\]</p>\n\nwhere \\(\\hat{y_{i}}\\) is the predicted relevance score for each product-query pair, and y is the actual relevance score assigned by human raters (i.e., the ground truth).\n\n<p>We want to minimize RMSE. Smaller values, towards 0, indicate better performance (because RMSE is always non-negative).</p>\n\n<h2>Submission Format</h2>\nFor each id (of the product-query pair) in the test set, you must predict a relevance score. .  The file should contain a header and have the following format:\n\n<p>A submission file should contain <strong>two columns</strong>: id (of the product-query pair) and the predicted relevance score.  Columns must be separated by a comma.  The file should must ontain a header line, and should have the following format:\n</p><pre>id,relevance\n1,1\n4,2\n5,3\n...\n</pre>\nThe relevance scores must be a real number in the range [1,3].",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}