{
  "fetched_at": "2026-10-19T08:54:59Z",
  "source_url": "https://www.kaggle.com/competitions/expedia-hotel-ranking/data",
  "pages": [
    {
      "name": "rules",
      "content": "One account per participant: You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n\nNo sharing of data: The data provided in this competition should be used solely for the purpose of this recruitment effort. Any sharing of data or code is not permitted. Reverse engineer anonymised column names after the competition has ended (for purposes beyond feature engineering) is not permitted.\n\nSubmission anonymity: If you would like your submission’s user name to be anonymous, please create a new Kaggle account for this recruitment effort.\n\nTeam Mergers: Team mergers are not allowed in this competition.\n\nSubmission: Limits You may submit a maximum of 20 entries per day.\n\nCompetition Timeline Start Date: You choose the date to start with the recruiter\n\nEnd Date: 48 hours after the agreed start date",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "In order to solve this exercise you will need the train and test csv files. train.csv contains ground-truth relevance features and the ground-truth relevance (derived from observing consumer's purchase decision) and needs to be used for model training, test.csv will be the basis for your submission.\n\nFile descriptions\n\n- a_train.csv - the training set\n- a_test.csv - the test set\n\nData fields\n\n- srch_id: unique identifier for a search\n- visitor_id: visitor id used to submit the search \n- visitor_visit_nbr: count of visits for the user who has submitted the search \n- visitor_loc_country: country of the user who has submitted the search \n- visitor_loc_region: region of the user who has submitted the search \n- visitor_loc_city: city of the user who has submitted the search \n- visitor_wr_member: indicates if the user who has submitted the search is part of the hcom loyalty program\n- posa_continent: point of sale continent for a partiular search\n- posa_country: point of sale country for a particular search\n- srch_hcom_destination_id: destination id for a particular search \n- srch_dest_longitude: destination center longitude for a particular search\n- srch_dest_latitude: destination center latitude for a particular search\n- srch_ci: date of check-in of a search\n- srch_co: date of cehck-in of a search\n- srch_ci_day: day of check-in of a search\n- srch_co_day: day of check-out of a search\n- srch_los: length of stay of a search (check-in - check-out date)\n- srch_bw: booking window of a search (search-date - check-in date)\n- srch_adults_cnt: number of adults for a search \n- srch_children_cnt: number of children for a search \n- srch_rm_cnt: number of requested rooms on the search \n- mobile_bool: indicates if the search was submitted on a mobile browser\n- mobile_app: indicates if the search was submitted on a mobile app\n- device: identifies the device type type used for this search\n- currency: currency relevant for the point of sale the search was submitted form\n- travelad_bool: indicates if a porperty in a search was a paid advert\n- dotd_bool: indicates if a property in a search was advertised as a special \"daily deal\"\n- price_without_discount_local: discounted property price in local (posa specific) currency\n- price_without_discount_usd: discounted property price in USD\n- price_with_discount_local: non-discounted property price in local currency\n- price_with_discount_usd: non-discounted property price in USD\n- imp_drr: name of the pricing/discount rule relevant for this property\n- booking_bool: indicates if property has been booken in a given search\n- prop_key: unique property identifier\n- prop_brand_bool: indicates if a property has a popular brand\n- prop_starrating: conventional hotel star rating (1-5 Stars)\n- prop_super_region: the super region this property belongs to (EMEA, APAC, NA, LATAM)\n- prop_continent: continent this property is located on (Europe, North-America, South-America, ..... )\n- prop_country: country this property is located in\n- prop_market_id: higher granularity geo classification\n- prop_submarket_id: highest granularity geo classification\n- prop_room_capacity: max number of rooms this property has available\n- prop_review_score: average guest review score for this property\n- prop_review_count: count of customer reviews for this property\n- prop_hostel_bool: indicates if a property is a hostel\n- local_date: date for a particular search",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "In this task, we ask you to train a model that learns to rank properties.\n\nHotels.com travellers provide their trip information, like destination, holiday dates and number of people in the search. Our search engine then returns a list of properties, ranked by their relevance to that traveller. \nWe provide you a data set with this travel information augmented by some features we have found to be predictive, including search-level features like booking window (length of time between booking the trip and taking it), and property-level features, like the country the property is in. \nPlease feel free to model with all, some or none of these, or create your own features.\nThe label in this data set is booking_bool, indicating whether this property was the one that was booked by the customer.\nPlease provide justification for your metric for this ranking task.\n\nIn the leader board, you’ll see the popularity benchmark. This is a useful baseline. It’s the score you’d achieve on the test set if you simply ranked the properties in each search by the number of times it was booked in the training set.\n\nHappy modelling, and good luck!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Evaluation Metric is NDCG@50. Relevance is given by propbookingbool\n(https://en.wikipedia.org/wiki/Discounted_cumulative_gain#Normalized_DCG)\n\nsubmission format \n\nSuppose that in our test file we only have two searches:\n\nsrchid 1234 with propkeys {546,678,678}\nsrchid 2345 with propkeys {987,849,839,930}\n\nsuppose that after scoring this queries you have produced the following scores:\n\nsrch\\_id, prop\\_key, score \n1234, 546, 10.3\n1234, 678, 8.3\n1234, 890, 3.0\n2345, 987, 2.0\n2345, 849, 1.0\n2345, 839, 3.0\n2345, 930, 10.0\n\nthe your submission should look like:\n\nsrch\\_id, prop\\_key\n1234, 546\n1234, 678\n1234, 890 \n2345, 930\n2345, 839\n2345, 987\n2345, 849",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}