{
  "fetched_at": "2026-10-19T10:40:33Z",
  "source_url": "https://www.kaggle.com/competitions/conversion-rate-prediction/data",
  "pages": [
    {
      "name": "data-description",
      "content": "\n\n## Files\n\n*   **train.csv** - the training set\n*   **test.csv** - the test set\n*   **sample_submission.csv** - a sample submission file in the correct format\n\nIn both, the training set and the test set, each row corresponds to a listing's impression at MercadoLibre. That means that the same item will appear many times on the data.\n\nThe `conversion` column contains a boolean that specifies whether an order was attributed to that impression or not. In other words, it tells us if the user who was presented with that listing decided to purchase the item.\n\n## Columns\n\nThe following columns are available on the training set: \n\n* `accepts_mercadopago`\tWhether the item accepts Mercado Pago\n* `available_quantity`\tThe available stock quantity at that moment\n*  `avg_asp_item_bday`\tAverage selling price of this item over the last days\n*  `avg_asp_item_domain`\tAverage price of items of the domain this item belongs to\n* `avg_asp_item_sel`\tAverage price of seller sales\n* `avg_gmv_item_bday`\taverage revenue generated  by the item per day\n* `avg_gmv_item_domain_30days`\tAverage revenue generated by the items of this domain on the last month\n* `avg_gmv_item_sel`\tAverage revenue of items of this seller\n* `avg_gmv_seller_bday`\tAverage revenue this seller makes by day\n* `avg_orders_item_bday`\tAverage number of orders this item has by day\n* `avg_orders_seller_bday`\tAverage orders the seller has by day\n* `avg_qty_orders_item_domain_30days`\tAverage number of orders a random item of this domain made on the last month\n* `avg_qty_orders_item_sel_30days`\tAverage number of orders an item of this seller makes on the last 30 days\n* `avg_si_item_bday`\tAverage units sold this item has by day\n* `avg_si_item_sel_30day`\tAverage units sold of an item of this seller on the past month\n* `avg_visits_item`\tAverage visits this item had\n* `benefit`\tIgnore, should be dropped\n* `boosted`\tWhether the item was boosted\n* `buy_server_timestamp`\tTimestamp of the purchase\n* `category_id`\tCategory of this item\n* `conversion`\tTarget variable, it is True if this print has an attributed order\n* `cus_cust_id`\tBuyer id\n* `cus_cust_id_sel`\tSeller id\n* `date`\tPrint date\n* `deal_print_id`\tUnique id for the print\n* `decimals`\tIgnore, will be dropped\n* `domain_id`\tDomain id for the item\n* `etl_version`\tIgnore, should be dropped\n* `free_shipping`\tWhether the item has free shipping\n* `fulfillment`\tWhether the item is fulfilled by MeLi\n* `full_name`\tCategory full name\n* `health`\tItem health\n* `is_pdp`\tWhether the click landed on a PDP\n* `product_id`\tProduct_id of the item\n* `item_id`\tID of the item, useful for debugging\n* `listing_type_id`\tWhether the item is gold or not\n* `logistic_type`\tLogistic type for the item\n* `main_picture`\tURL for the main item picture\n* `offset`\tOn which page the item was rendered\n* `original_price`\tPrice from which the discount was done\n* `platform`\tWhich platform the user is using\n* `price`\tItem price\n* `print_position`\tPosition on the page\n* `print_server_timestamp`\tTimestamp for the print\n* `qty_items_dom`\tNumber of items this domain has\n* `qty_items_sel`\tNumber of items the seller has\n* `rn`\tLeftover from the ETL, Discard\n* `site_id`\tSite ID\n* `sold_quantity`\tNumber of items sold at the moment of the print\n* `tags`\tTags the item had at the moment of the print\n* `title`\tItem title\n* `total_asp_item_domain_30days`\tAverage selling price of the items of the domain\n* `total_asp_item_sel_30days`\tAverage selling price of all the items the seller sold on the last 30 days\n* `total_gmv_domain_30days`\tTotal revenue the domain made on the last 30 days\n* `total_gmv_domain_bday`\ttotal_gmv_domain_30days / 30\n* `total_gmv_item_30days`\tTotal revenue made by the item on the lasts 30 days\n* `total_gmv_seller`\tTotal revenue made by the seller on the last month\n* `total_items_domain`\tNumber of items on the domain\n* `total_items_seller`\tNumber of items the seller has\n* `total_orders_domain_30days`\tTotal orders on the domain\n* `total_orders_item_30days`\tTotal orders the Item had on the last 30 days\n* `total_orders_sel_30days`\tTotal orders for the seller\n* `total_si_domain_30days`\tTotal units sold of this domain\n* `total_si_item_30days`\tTotal units sold of this item\n* `total_si_sel_30days`\tSame for the seller\n* `total_visits_domain`\tTotal visits on this domain\n* `total_visits_item`\tTotal visits this item had\n* `total_visits_seller`\tTotal visits for this seller\n* `uid`\tsession id\n* `user_id`\tuser id\n* `warranty`\tWhether the item had warranty\n\nThe test set differs from the training set on just two columns: It contains a `ROW_ID` and it **does not** contain the `conversion` column for obvious reasons. The `ROW_ID` will be useful to identify each sample in your submission file.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "This competition takes place in the context of the course \"Introducción a Ciencia de Datos\" at Universidad Torcuato Di Tella. \n\nThe objective of this challenge is to predict conversion for an online retailer based on the item features. E-commerce conversions are defined as the number of times that users complete a desired action on the platform. In this case, the desired action is to purchase a product. \n\nBeing able to estimate conversion rate is critical for an online retailer as it impacts many aspects of the business, for example: the quality of the recommendations, search results and product ads they display to their users. The same user may or may not purchase the product under different circumstances.\n\n\n## Acknowledgements\n\nThe data was made publicly available by MercadoLibre in the context of #MeliDataChallenge2019",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Submissions will be scored using the AUC ROC score. For more information regarding this metric refer to your class notes.\n\n## Submission Format\n\n**For every participant**, submission files should be `.csv` formatted and contain two columns: `ROW_ID` and `conversion`\n\n\nThe file should contain a header and have the following format:\n```\nROW_ID,conversion\n(string),(double)\n```",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "### One account per participant\nYou cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n\n### No private sharing outside teams\nPrivately sharing code or data outside of teams is not permitted.\nIt's okay to share code if made available to all participants on the forums.\n\n### Discussion\nPlease use the forum to post any question you may have during the course of the challenge. All participants are invited to respond. Professors will also respond on the forum so that everyone benefits from the discussion. \n\n### Teams\nTeams are expected to be conformed in the same way as the ones registered in the course.\nTeam mergers are not allowed in this competition.\n\n### Submission Limits:\nYou may submit a maximum of 5 entries per day.\nYou may select only 1 final submissions for judging.\n\n<h3>Competition Timeline</h3>\n<p>Start Date: <strong>24/10/2021</strong></p>\n<p>Entry Deadline: <strong> All teams must have submitted once before  05/11/2021 </strong></p>\n<p>End Date: <strong>19/11/2021 12:00 AM UTC</strong></p>\n\n### Do not forget \n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}