Can one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?
\n\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).
", "messages": [ { "id": "116399", "postDate": "04/23/2016 22:44:58", "content": "Can one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?
\n\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).
", "rawMarkdown": "Can one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?\r\n\r\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).", "votes": null }, { "id": "116456", "postDate": "04/24/2016 13:18:57", "content": "I suppose those are not cluster changes, but the user is actually checking different clusters before making a final booking decision.
", "rawMarkdown": "I suppose those are not cluster changes, but the user is actually checking different clusters before making a final booking decision.", "votes": null }, { "id": "116599", "postDate": "04/25/2016 11:09:40", "content": "[quote=Marcin Pękalski;116399]
\n\nCan one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?
\n\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).
\n\n[/quote]
\n\nNo, we don't take room types and services into account. Clustering changing 4 times within 3 minutes most likely means that a user if checking different hotels.
", "rawMarkdown": "[quote=Marcin Pękalski;116399]\r\n\r\nCan one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?\r\n\r\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).\r\n\r\n[/quote]\r\n\r\nNo, we don't take room types and services into account. Clustering changing 4 times within 3 minutes most likely means that a user if checking different hotels.", "votes": null }, { "id": "116606", "postDate": "04/25/2016 11:36:00", "content": "One more thing to keep in mind: clustering is also based on distance of a hotel from a center of srch_destination_id. What it means is that clustering assignment of a hotel e.g. in New York can change depending on whether it was clicked/booked from "New York and vicinity" or "Statue of Liberty".
", "rawMarkdown": "One more thing to keep in mind: clustering is also based on distance of a hotel from a center of srch_destination_id. What it means is that clustering assignment of a hotel e.g. in New York can change depending on whether it was clicked/booked from \"New York and vicinity\" or \"Statue of Liberty\".", "votes": null }, { "id": "116915", "postDate": "04/26/2016 14:40:01", "content": "Some lines from train:
\n\n['2014-10-28 15:02:03', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '91']\n['2014-10-28 15:09:58', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '94']\nThat's really strange. ) Probably some data is missing in CSV to predict exact cluster.
", "rawMarkdown": "Some lines from train:\r\n\r\n ['2014-10-28 15:02:03', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '91']\r\n ['2014-10-28 15:09:58', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '94']\r\n\r\nThat's really strange. ) Probably some data is missing in CSV to predict exact cluster.", "votes": null }, { "id": "116918", "postDate": "04/26/2016 14:52:49", "content": "Why are these lines strange? I think it simply means that a user clicked two different hotels...
", "rawMarkdown": "Why are these lines strange? I think it simply means that a user clicked two different hotels...", "votes": null }, { "id": "116919", "postDate": "04/26/2016 14:55:24", "content": "The distance from the user to the hotel is identical in both: 0.0056, what suggests it's the exact same hotel.
", "rawMarkdown": "The distance from the user to the hotel is identical in both: 0.0056, what suggests it's the exact same hotel.", "votes": null }, { "id": "116920", "postDate": "04/26/2016 14:59:10", "content": "[quote=Adam;116918]
\n\nWhy are these lines strange? I think it simply means that a user clicked two different hotels...
\n\n[/quote]
\n\nIt makes sense in case hotels located on the same distance from user. Thank you.
", "rawMarkdown": "[quote=Adam;116918]\r\n\r\nWhy are these lines strange? I think it simply means that a user clicked two different hotels...\r\n\r\n[/quote]\r\n\r\nIt makes sense in case hotels located on the same distance from user. Thank you.", "votes": null }, { "id": "116924", "postDate": "04/26/2016 15:06:45", "content": "I've double checked raw data and these two records correspond to different hotels.
\n\nKeep in mind that orig_destination_distance is rounded to significant 4 digits after comma, and many destinations are dense with hotels, that's why it's not uncommon to see the same value of orig_destination_distance for different hotels.
", "rawMarkdown": "I've double checked raw data and these two records correspond to different hotels. \r\n\r\nKeep in mind that orig_destination_distance is rounded to significant 4 digits after comma, and many destinations are dense with hotels, that's why it's not uncommon to see the same value of orig_destination_distance for different hotels.", "votes": null }, { "id": "116925", "postDate": "04/26/2016 15:08:29", "content": "What's the unit for orig_destination_distance?
", "rawMarkdown": "What's the unit for orig_destination_distance?", "votes": null }, { "id": "116926", "postDate": "04/26/2016 15:09:48", "content": "miles
", "rawMarkdown": "miles", "votes": null } ], "comments": [ { "id": 116456, "author_name": "hitoshinagano", "author_url": "", "post_date": "04/24/2016 13:18:57", "content": "I suppose those are not cluster changes, but the user is actually checking different clusters before making a final booking decision.
", "votes": null, "replies": [] }, { "id": 116599, "author_name": "adamwoz", "author_url": "", "post_date": "04/25/2016 11:09:40", "content": "[quote=Marcin Pękalski;116399]
\n\nCan one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?
\n\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).
\n\n[/quote]
\n\nNo, we don't take room types and services into account. Clustering changing 4 times within 3 minutes most likely means that a user if checking different hotels.
", "votes": null, "replies": [] }, { "id": 116606, "author_name": "adamwoz", "author_url": "", "post_date": "04/25/2016 11:36:00", "content": "One more thing to keep in mind: clustering is also based on distance of a hotel from a center of srch_destination_id. What it means is that clustering assignment of a hotel e.g. in New York can change depending on whether it was clicked/booked from "New York and vicinity" or "Statue of Liberty".
", "votes": null, "replies": [] }, { "id": 116915, "author_name": "zfturbo", "author_url": "", "post_date": "04/26/2016 14:40:01", "content": "Some lines from train:
\n\n['2014-10-28 15:02:03', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '91']\n['2014-10-28 15:09:58', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '94']\nThat's really strange. ) Probably some data is missing in CSV to predict exact cluster.
", "votes": null, "replies": [] }, { "id": 116918, "author_name": "adamwoz", "author_url": "", "post_date": "04/26/2016 14:52:49", "content": "Why are these lines strange? I think it simply means that a user clicked two different hotels...
", "votes": null, "replies": [] }, { "id": 116919, "author_name": "radek1st", "author_url": "", "post_date": "04/26/2016 14:55:24", "content": "The distance from the user to the hotel is identical in both: 0.0056, what suggests it's the exact same hotel.
", "votes": null, "replies": [] }, { "id": 116920, "author_name": "zfturbo", "author_url": "", "post_date": "04/26/2016 14:59:10", "content": "[quote=Adam;116918]
\n\nWhy are these lines strange? I think it simply means that a user clicked two different hotels...
\n\n[/quote]
\n\nIt makes sense in case hotels located on the same distance from user. Thank you.
", "votes": null, "replies": [] }, { "id": 116924, "author_name": "adamwoz", "author_url": "", "post_date": "04/26/2016 15:06:45", "content": "I've double checked raw data and these two records correspond to different hotels.
\n\nKeep in mind that orig_destination_distance is rounded to significant 4 digits after comma, and many destinations are dense with hotels, that's why it's not uncommon to see the same value of orig_destination_distance for different hotels.
", "votes": null, "replies": [] }, { "id": 116925, "author_name": "radek1st", "author_url": "", "post_date": "04/26/2016 15:08:29", "content": "What's the unit for orig_destination_distance?
", "votes": null, "replies": [] }, { "id": 116926, "author_name": "adamwoz", "author_url": "", "post_date": "04/26/2016 15:09:48", "content": "miles
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "116399": "Can one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?\r\n\r\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).", "116456": "I suppose those are not cluster changes, but the user is actually checking different clusters before making a final booking decision.", "116599": "[quote=Marcin Pękalski;116399]\r\n\r\nCan one hotel belong at the same time to several hotel_clusters depending on, for example, type of room picked by the user or services included? So in reality we are trying to predict what services will customer prefer?\r\n\r\nOtherwise I somehow cannot understand how clustering can change 4 times within 3 minutes (being browsed from the same site_name).\r\n\r\n[/quote]\r\n\r\nNo, we don't take room types and services into account. Clustering changing 4 times within 3 minutes most likely means that a user if checking different hotels.", "116606": "One more thing to keep in mind: clustering is also based on distance of a hotel from a center of srch_destination_id. What it means is that clustering assignment of a hotel e.g. in New York can change depending on whether it was clicked/booked from \"New York and vicinity\" or \"Statue of Liberty\".", "116915": "Some lines from train:\r\n\r\n ['2014-10-28 15:02:03', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '91']\r\n ['2014-10-28 15:09:58', '2', '3', '66', '174', '23702', '0.0056', '551586', '0', '0', '9', '2014-11-02', '2014-11-07', '1', '0', '1', '5797', '3', '0', '1', '2', '50', '366', '94']\r\n\r\nThat's really strange. ) Probably some data is missing in CSV to predict exact cluster.", "116918": "Why are these lines strange? I think it simply means that a user clicked two different hotels...", "116919": "The distance from the user to the hotel is identical in both: 0.0056, what suggests it's the exact same hotel.", "116920": "[quote=Adam;116918]\r\n\r\nWhy are these lines strange? I think it simply means that a user clicked two different hotels...\r\n\r\n[/quote]\r\n\r\nIt makes sense in case hotels located on the same distance from user. Thank you.", "116924": "I've double checked raw data and these two records correspond to different hotels. \r\n\r\nKeep in mind that orig_destination_distance is rounded to significant 4 digits after comma, and many destinations are dense with hotels, that's why it's not uncommon to see the same value of orig_destination_distance for different hotels.", "116925": "What's the unit for orig_destination_distance?", "116926": "miles" }, "source": "meta" }