Google analytics has the option to create a Demo account with data from Google Merchandise Store, which should be the same used in this competition. Informations are similar to this dataset, but the revenue for the train set period (Aug 2016 to Aug 2017) is around 4.5M in Demo account, while here it is 1.5M. The number of sessions is also higher (970k vs 903k). Does someone know the difference between the dataset we have avaliable and the one used in Demo account?
", "messages": [ { "id": 401395, "postDate": "2018-10-10T02:30:40.740Z", "content": "Google analytics has the option to create a Demo account with data from Google Merchandise Store, which should be the same used in this competition. Informations are similar to this dataset, but the revenue for the train set period (Aug 2016 to Aug 2017) is around 4.5M in Demo account, while here it is 1.5M. The number of sessions is also higher (970k vs 903k). Does someone know the difference between the dataset we have avaliable and the one used in Demo account?
", "rawMarkdown": "Google analytics has the option to create a Demo account with data from Google Merchandise Store, which should be the same used in this competition. Informations are similar to this dataset, but the revenue for the train set period (Aug 2016 to Aug 2017) is around 4.5M in Demo account, while here it is 1.5M. The number of sessions is also higher (970k vs 903k). Does someone know the difference between the dataset we have avaliable and the one used in Demo account?", "votes": 13 }, { "id": 401458, "postDate": "2018-10-10T05:41:26.263Z", "content": "I don't know what is the difference. But I think that the modifications made by Kaggle to the original dataset will not be enough to save this competition, whose days are unfortunately now numbered...
", "rawMarkdown": "I don't know what is the difference. But I think that the modifications made by Kaggle to the original dataset will not be enough to save this competition, whose days are unfortunately now numbered...", "votes": 11, "replies": [ { "id": 401809, "postDate": "2018-10-10T18:16:45.257Z", "content": "@Jerome
\n\nFully agree with you.
", "rawMarkdown": "@Jerome \n\nFully agree with you.", "votes": 2 } ] }, { "id": 402402, "postDate": "2018-10-11T16:43:50.470Z", "content": "I purposely did not look into this. But am not surprised at all. I think I was the first one using the demo account......
", "rawMarkdown": "I purposely did not look into this. But am not surprised at all. I think I was the first one using the demo account......", "votes": 1 }, { "id": 401637, "postDate": "2018-10-10T11:55:25.260Z", "content": "I guess we're gonna see quite a few more \"I can see the future\" kernels some time soon... I'm a bit sad that this external data is available...
", "rawMarkdown": "I guess we're gonna see quite a few more \"I can see the future\" kernels some time soon... I'm a bit sad that this external data is available... ", "votes": 1, "replies": [ { "id": 401643, "postDate": "2018-10-10T12:16:01.090Z", "content": "Is the fullVisitorID in this external data the same with competition data for a specific user?
", "rawMarkdown": "Is the fullVisitorID in this external data the same with competition data for a specific user?" }, { "id": 401644, "postDate": "2018-10-10T12:19:38.393Z", "content": "I haven't seen fullVisitorID data yet, but so far I've been able to compare stats by category for the train vs test period. Even if our data is a subset of the whole store, I can't see how this isn't gonna be a big leak.
", "rawMarkdown": "I haven't seen fullVisitorID data yet, but so far I've been able to compare stats by category for the train vs test period. Even if our data is a subset of the whole store, I can't see how this isn't gonna be a big leak." }, { "id": 401646, "postDate": "2018-10-10T12:26:15.567Z", "rawMarkdown": "", "isDeleted": true }, { "id": 401725, "postDate": "2018-10-10T15:36:45.457Z", "content": "The problem is this gap between revenue (almost 3 times). Even if the GA information is not usefull, it would be important to know how it is related, since one of them has to be the real data and the other a subsample or fake dataset...
", "rawMarkdown": "The problem is this gap between revenue (almost 3 times). Even if the GA information is not usefull, it would be important to know how it is related, since one of them has to be the real data and the other a subsample or fake dataset...", "votes": 2 } ] }, { "id": 401831, "postDate": "2018-10-10T18:52:56.933Z", "content": "I also believe there are visitors that have been removed completely and I've seen one for whom all sessions but one have been removed. So we have a visitor that bounces and never comes back instead of being a returning visitor and purchasing customer.
", "rawMarkdown": "I also believe there are visitors that have been removed completely and I've seen one for whom all sessions but one have been removed. So we have a visitor that bounces and never comes back instead of being a returning visitor and purchasing customer." }, { "id": 401827, "postDate": "2018-10-10T18:49:35.600Z", "content": "After taking a look, I think there's a leak. I haven't been able to confirm yet, but will try again when I have more time and submissions available.
", "rawMarkdown": "After taking a look, I think there's a leak. I haven't been able to confirm yet, but will try again when I have more time and submissions available." } ], "comments": [ { "id": 401458, "author_name": "Jerome Vallet", "author_url": "", "post_date": "2018-10-10T05:41:26.263000", "content": "I don't know what is the difference. But I think that the modifications made by Kaggle to the original dataset will not be enough to save this competition, whose days are unfortunately now numbered...
", "votes": 11, "replies": [ { "id": 401809, "author_name": "mezoganet", "author_url": "", "post_date": "2018-10-10T18:16:45.257000", "content": "@Jerome
\n\nFully agree with you.
", "votes": 2, "replies": [] } ] }, { "id": 402402, "author_name": "Erik Bruin", "author_url": "", "post_date": "2018-10-11T16:43:50.470000", "content": "I purposely did not look into this. But am not surprised at all. I think I was the first one using the demo account......
", "votes": 1, "replies": [] }, { "id": 401637, "author_name": "S D", "author_url": "", "post_date": "2018-10-10T11:55:25.260000", "content": "I guess we're gonna see quite a few more \"I can see the future\" kernels some time soon... I'm a bit sad that this external data is available...
", "votes": 1, "replies": [ { "id": 401643, "author_name": "fortune cookie", "author_url": "", "post_date": "2018-10-10T12:16:01.090000", "content": "Is the fullVisitorID in this external data the same with competition data for a specific user?
", "votes": 0, "replies": [] }, { "id": 401644, "author_name": "S D", "author_url": "", "post_date": "2018-10-10T12:19:38.393000", "content": "I haven't seen fullVisitorID data yet, but so far I've been able to compare stats by category for the train vs test period. Even if our data is a subset of the whole store, I can't see how this isn't gonna be a big leak.
", "votes": 0, "replies": [] }, { "id": 401646, "author_name": "", "author_url": "", "post_date": "2018-10-10T12:26:15.567000", "content": "", "votes": 0, "replies": [] }, { "id": 401725, "author_name": "Aguiar", "author_url": "", "post_date": "2018-10-10T15:36:45.457000", "content": "The problem is this gap between revenue (almost 3 times). Even if the GA information is not usefull, it would be important to know how it is related, since one of them has to be the real data and the other a subsample or fake dataset...
", "votes": 2, "replies": [] } ] }, { "id": 401831, "author_name": "S D", "author_url": "", "post_date": "2018-10-10T18:52:56.933000", "content": "I also believe there are visitors that have been removed completely and I've seen one for whom all sessions but one have been removed. So we have a visitor that bounces and never comes back instead of being a returning visitor and purchasing customer.
", "votes": 0, "replies": [] }, { "id": 401827, "author_name": "S D", "author_url": "", "post_date": "2018-10-10T18:49:35.600000", "content": "After taking a look, I think there's a leak. I haven't been able to confirm yet, but will try again when I have more time and submissions available.
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "401395": "Google analytics has the option to create a Demo account with data from Google Merchandise Store, which should be the same used in this competition. Informations are similar to this dataset, but the revenue for the train set period (Aug 2016 to Aug 2017) is around 4.5M in Demo account, while here it is 1.5M. The number of sessions is also higher (970k vs 903k). Does someone know the difference between the dataset we have avaliable and the one used in Demo account?", "401458": "I don't know what is the difference. But I think that the modifications made by Kaggle to the original dataset will not be enough to save this competition, whose days are unfortunately now numbered...", "402402": "I purposely did not look into this. But am not surprised at all. I think I was the first one using the demo account......", "401637": "I guess we're gonna see quite a few more \"I can see the future\" kernels some time soon... I'm a bit sad that this external data is available... ", "401831": "I also believe there are visitors that have been removed completely and I've seen one for whom all sessions but one have been removed. So we have a visitor that bounces and never comes back instead of being a returning visitor and purchasing customer.", "401827": "After taking a look, I think there's a leak. I haven't been able to confirm yet, but will try again when I have more time and submissions available." } }