{
  "id": 10729,
  "title": "Welcome to Avazu CTR Prediction Contest!",
  "url": "/competitions/avazu-ctr-prediction/discussion/10729",
  "author_name": "",
  "post_date": "2014-10-25T02:21:52.640Z",
  "votes": 21,
  "comment_count": 31,
  "views": 18358,
  "content": "<p>Hello Everybody,</p>\n\n<p>Welcome! This is Steve, product manager from <a href=\"http://avazuinc.com/\" target=\"_blank\">Avazu</a>. <a href=\"http://avazuinc.com/\" target=\"_blank\">Avazu (http://avazuinc.com/)</a>&nbsp;is&nbsp;a leading multinational corporation in the digital marketing industry, <br>specializing in cross-device advertising and mobile game publishing.</p>\n<p>Digital advertising world is quickly&nbsp;evolving, which is why Avazu has always been investing more in R&amp;D. CTR prediction is one important step of targeting technology.</p>\n\n<p>Today we have released&nbsp;11 days worth of Avazu data (mobile ads) to build and test prediction models. Your task is quite simple: predicting whether a mobile ad will be clicked or not. Can you find a strategy that beats standard classification algorithms?</p>\n\n<p>We have also prepared great prizes ($15,000 in total) for the winners, and the first prize winner will get $10,000!<br><br></p>\n<p>Exciting! Isn't it?</p>\n\n<p>Why not just come and join this great competition!</p>\n<p>I am always happy to help&nbsp;you.&nbsp;Just hit me with your questions in this forum.</p>\n\n<p>Cheers!&nbsp;</p>",
  "messages": [
    {
      "id": "56703",
      "postDate": "10/25/2014 02:21:52",
      "content": "<p>Hello Everybody,</p>\n\n<p>Welcome! This is Steve, product manager from <a href=\"http://avazuinc.com/\" target=\"_blank\">Avazu</a>. <a href=\"http://avazuinc.com/\" target=\"_blank\">Avazu (http://avazuinc.com/)</a>&nbsp;is&nbsp;a leading multinational corporation in the digital marketing industry, <br>specializing in cross-device advertising and mobile game publishing.</p>\n<p>Digital advertising world is quickly&nbsp;evolving, which is why Avazu has always been investing more in R&amp;D. CTR prediction is one important step of targeting technology.</p>\n\n<p>Today we have released&nbsp;11 days worth of Avazu data (mobile ads) to build and test prediction models. Your task is quite simple: predicting whether a mobile ad will be clicked or not. Can you find a strategy that beats standard classification algorithms?</p>\n\n<p>We have also prepared great prizes ($15,000 in total) for the winners, and the first prize winner will get $10,000!<br><br></p>\n<p>Exciting! Isn't it?</p>\n\n<p>Why not just come and join this great competition!</p>\n<p>I am always happy to help&nbsp;you.&nbsp;Just hit me with your questions in this forum.</p>\n\n<p>Cheers!&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56704",
      "postDate": "10/25/2014 02:22:34",
      "content": "<p>cheers!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56805",
      "postDate": "10/26/2014 16:47:23",
      "content": "<p>Can the organizers provide some sense of the type of data represented by these anonymized variables? A mix of description about the ad and the recipient? Can any information be supplied?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56818",
      "postDate": "10/27/2014 03:26:40",
      "content": "<p>the meanings of some variables (starting from 5th to 19th variable) are</p>\n<p>,banner_pos<br> ,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model <br> ,device_type<br> ,device_conn_type<br> ,device_geo_country</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56820",
      "postDate": "10/27/2014 03:38:06",
      "content": "<p>Thanks for providing this. &nbsp;It's much more fun when you're able to use some intuition in model building.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56823",
      "postDate": "10/27/2014 05:27:22",
      "content": "<p>I can't find where to download the data set, anyone can tell the links. Thanks!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56825",
      "postDate": "10/27/2014 05:54:32",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>the meanings of some anonymized categorical variables (starting from 6th to 16th variable) are</p>\n<p>,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model</p>\n<p>[/quote]</p>\n\n<p>That's great and already gives me some idea about what to do with these features, but why only 6 through 16? I think if you want us to squeeze some info out, you'd do best to just tell us what they all are, if possible.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56833",
      "postDate": "10/27/2014 06:30:21",
      "content": "<p>[quote=Phillip Chilton Adkins;56825]</p>\n<p>That's great and already gives me some idea about what to do with these features, but why only 6 through 16? I think if you want us to squeeze some info out, you'd do best to just tell us what they all are, if possible.</p>\n<p>[/quote]</p>\n<p>Hi&nbsp;Phillip</p>\n<p>I have edited the post and provided more variable meanings</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56834",
      "postDate": "10/27/2014 07:36:45",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>the meanings of some variables (starting from 5th to 19th variable) are</p>\n<p>,banner_pos<br> ,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model <br> ,device_type<br> ,device_conn_type<br> ,device_geo_country</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>In my opinion, this information is very relevant and belongs to the &quot;data&quot; page, why make participants dig for it on a forum?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56872",
      "postDate": "10/28/2014 00:13:35",
      "content": "<p>[quote=Foxtrot;56834]</p>\n<p>In my opinion, this information is very relevant and belongs to the &quot;data&quot; page, why make participants dig for it on a forum?</p>\n<p>[/quote]</p>\n\n<p>100% agreed, sirs</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "56873",
      "postDate": "10/28/2014 00:15:09",
      "content": "<p>These column headers will appear on the data files. You will not need to hunt for them&nbsp;in the forums. Please stay tuned for when data downloads are re-enabled. Thanks.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57272",
      "postDate": "11/04/2014 05:20:54",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>...</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>Do you mean C1 and C17 to C24 are all categorical variables ?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57273",
      "postDate": "11/04/2014 05:51:44",
      "content": "<p>[quote=B Yang;57272]</p>\n<p>[quote=Steve Wang;56818]</p>\n<p>...</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>Do you mean C1 and C17 to C24 are all categorical variables ?</p>\n<p>[/quote]</p>\n<p>yes</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57366",
      "postDate": "11/06/2014 00:12:30",
      "content": "<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57373",
      "postDate": "11/06/2014 03:16:54",
      "content": "<p>[quote=Payam;57366]</p>\n<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>\n<p>[/quote]</p>\n\n<p>Hello, we have informed our Kaggle admin to do the editing. Thanks!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57381",
      "postDate": "11/06/2014 05:06:21",
      "content": "<p>[quote=Payam;57366]</p>\n<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>\n<p>[/quote]</p>\n<p>It does not matter if someone is expert or beginner; missing data types causes unnecessary problems for everyone. It simply make no sense to guess these meta data in anyway. Who knows if any of the assumptions still holds for test data or not?</p>\n<p>I did suspect that some of the variables is ordinal, and that's a relevant piece of information.</p>\n<p>By the way, could the admin explain a little bit more about `device_make`, `device_model` and `device_type`? I can't really understand what do they mean exactly.</p>\n<p>And is there any&nbsp;hierarchical information in the features (like app_category&nbsp;&gt; app_domain &gt; app_id)? That could be relevant too, like once&nbsp;Y. Koren used taxonomy information of predicting music rating in&nbsp;KDDCup 2011.</p>\n<p>Ref:</p>\n<p>http://www.eng.tau.ac.il/~noamk/papers/DKK11.pdf</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57389",
      "postDate": "11/06/2014 07:15:41",
      "content": "<p>[quote=byronyi;57381]</p>\n<p>By the way, could the admin explain a little bit more about `device_make`, `device_model` and `device_type`? I can't really understand what do they mean exactly.</p>\n<p>[/quote]</p>\n<p>Make and model--if admins follow similar language to automobiles--is manufacturer and specific device. E.g. (make)(model) = (honda)(civic), or in our case (apple)(iphone6).</p>\n<p>I'd like some clarification on &quot;device_type&quot; too... it's smart phone/tablet/pc/etc maybe?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "57394",
      "postDate": "11/06/2014 09:39:50",
      "content": "<p>just say Hi!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "58294",
      "postDate": "11/18/2014 10:04:02",
      "content": "<p>Hi</p>\n<p>Could anyone let me know when Data downloads would be available?</p>\n<p>I am getting below message when click on Data Section.</p>\n<p>&quot;Update - Data downloads are temporarily disabled. Please see&nbsp;the forums&nbsp;for updates.&quot;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "58299",
      "postDate": "11/18/2014 10:39:49",
      "content": "<p>Thanks for the updates !</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "58682",
      "postDate": "11/22/2014 14:13:52",
      "content": "<p>Hello all,</p>\n<p>I've only been around since the new data set was released. In the old posts on this thread, there seems to be more info on what the different fields are. Concretely, <strong>device_os,</strong> <strong>device_make,</strong> and <strong>device_geo_country&nbsp;</strong>are gone in the new dataset description. Is this on purpose? Are they still there, behind C1 or C14-C21?</p>\n<p>Thanks!</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "58853",
      "postDate": "11/25/2014 05:59:02",
      "content": "<p>why the data have no ad id?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59075",
      "postDate": "11/27/2014 08:30:22",
      "content": "<p>Data in order to predict whether the advertisement is to click.However, we have no advertising information. Could you tell me whether the first field data are wrong.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59077",
      "postDate": "11/27/2014 08:35:02",
      "content": "<p>I didn't find any 2 identical ID advertising</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "59309",
      "postDate": "12/01/2014 06:21:21",
      "content": "<p>Hi Steve,</p>\n<p>Could you please tell me whether we need to include C1 and C15 - C24 in the analysis as they are just id's and&nbsp;did not give any description for these id's.</p>\n<p>Please clarify.</p>\n\n<p>Thanks,</p>\n<p>Sandeep</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "60132",
      "postDate": "12/13/2014 04:44:10",
      "content": "<p>hi all,</p>\n<p>what is the meaning of this feild ?</p>\n<p>'banner_pos'</p>\n<p>thank !</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "236361",
      "postDate": "10/27/2017 05:44:57",
      "content": "<p>Hi Sandeep, If you happened to know the answer for this kindly share </p>",
      "rawMarkdown": "Hi Sandeep, If you happened to know the answer for this kindly share",
      "votes": null
    },
    {
      "id": "969717",
      "postDate": "08/13/2020 21:59:30",
      "content": "<p><a href=\"https://www.kaggle.com/stevewang0\" target=\"_blank\">@stevewang0</a> , can I use this dataset for academic publication? thanks.</p>",
      "rawMarkdown": "stevewang0 , can I use this dataset for academic publication? thanks.",
      "votes": null
    },
    {
      "id": "1877012",
      "postDate": "07/30/2022 09:08:54",
      "content": "<p>Thank you for sharing</p>",
      "rawMarkdown": "Thank you for sharing",
      "votes": null
    },
    {
      "id": "2071829",
      "postDate": "12/21/2022 12:39:49",
      "content": "<p>Thank you for sharing :)</p>",
      "rawMarkdown": "Thank you for sharing :)",
      "votes": null
    },
    {
      "id": "2188128",
      "postDate": "03/19/2023 10:34:23",
      "content": "<p>Hello,<br>\nI wish to download datasets for CTR prediction for a research work. can you guide me here</p>",
      "rawMarkdown": "Hello,\nI wish to download datasets for CTR prediction for a research work. can you guide me here",
      "votes": null
    },
    {
      "id": "2563345",
      "postDate": "12/16/2023 08:05:50",
      "content": "<p>Thanks for the data that makes me a machine learning player.</p>",
      "rawMarkdown": "Thanks for the data that makes me a machine learning player.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 969717,
      "author_name": "julie0408",
      "author_url": "",
      "post_date": "08/13/2020 21:59:30",
      "content": "<p><a href=\"https://www.kaggle.com/stevewang0\" target=\"_blank\">@stevewang0</a> , can I use this dataset for academic publication? thanks.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1877012,
      "author_name": "matsudasatuki",
      "author_url": "",
      "post_date": "07/30/2022 09:08:54",
      "content": "<p>Thank you for sharing</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2071829,
      "author_name": "esh621",
      "author_url": "",
      "post_date": "12/21/2022 12:39:49",
      "content": "<p>Thank you for sharing :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2188128,
      "author_name": "lakshmidevirade",
      "author_url": "",
      "post_date": "03/19/2023 10:34:23",
      "content": "<p>Hello,<br>\nI wish to download datasets for CTR prediction for a research work. can you guide me here</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2563345,
      "author_name": "ericmsq",
      "author_url": "",
      "post_date": "12/16/2023 08:05:50",
      "content": "<p>Thanks for the data that makes me a machine learning player.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56704,
      "author_name": "jiweiliu",
      "author_url": "",
      "post_date": "10/25/2014 02:22:34",
      "content": "<p>cheers!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56805,
      "author_name": "inspector",
      "author_url": "",
      "post_date": "10/26/2014 16:47:23",
      "content": "<p>Can the organizers provide some sense of the type of data represented by these anonymized variables? A mix of description about the ad and the recipient? Can any information be supplied?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56818,
      "author_name": "stevewang0",
      "author_url": "",
      "post_date": "10/27/2014 03:26:40",
      "content": "<p>the meanings of some variables (starting from 5th to 19th variable) are</p>\n<p>,banner_pos<br> ,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model <br> ,device_type<br> ,device_conn_type<br> ,device_geo_country</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56820,
      "author_name": "dmitriyguller",
      "author_url": "",
      "post_date": "10/27/2014 03:38:06",
      "content": "<p>Thanks for providing this. &nbsp;It's much more fun when you're able to use some intuition in model building.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56823,
      "author_name": "xiximydream",
      "author_url": "",
      "post_date": "10/27/2014 05:27:22",
      "content": "<p>I can't find where to download the data set, anyone can tell the links. Thanks!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56825,
      "author_name": "phillipadkins",
      "author_url": "",
      "post_date": "10/27/2014 05:54:32",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>the meanings of some anonymized categorical variables (starting from 6th to 16th variable) are</p>\n<p>,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model</p>\n<p>[/quote]</p>\n\n<p>That's great and already gives me some idea about what to do with these features, but why only 6 through 16? I think if you want us to squeeze some info out, you'd do best to just tell us what they all are, if possible.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56833,
      "author_name": "stevewang0",
      "author_url": "",
      "post_date": "10/27/2014 06:30:21",
      "content": "<p>[quote=Phillip Chilton Adkins;56825]</p>\n<p>That's great and already gives me some idea about what to do with these features, but why only 6 through 16? I think if you want us to squeeze some info out, you'd do best to just tell us what they all are, if possible.</p>\n<p>[/quote]</p>\n<p>Hi&nbsp;Phillip</p>\n<p>I have edited the post and provided more variable meanings</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56834,
      "author_name": "zygmunt",
      "author_url": "",
      "post_date": "10/27/2014 07:36:45",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>the meanings of some variables (starting from 5th to 19th variable) are</p>\n<p>,banner_pos<br> ,site_id <br> ,site_domain <br> ,site_category<br> ,app_id <br> ,app_domain <br> ,app_category <br> ,device_id <br> ,device_ip <br> ,device_os <br> ,device_make <br> ,device_model <br> ,device_type<br> ,device_conn_type<br> ,device_geo_country</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>In my opinion, this information is very relevant and belongs to the &quot;data&quot; page, why make participants dig for it on a forum?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56872,
      "author_name": "phillipadkins",
      "author_url": "",
      "post_date": "10/28/2014 00:13:35",
      "content": "<p>[quote=Foxtrot;56834]</p>\n<p>In my opinion, this information is very relevant and belongs to the &quot;data&quot; page, why make participants dig for it on a forum?</p>\n<p>[/quote]</p>\n\n<p>100% agreed, sirs</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 56873,
      "author_name": "joycenv",
      "author_url": "",
      "post_date": "10/28/2014 00:15:09",
      "content": "<p>These column headers will appear on the data files. You will not need to hunt for them&nbsp;in the forums. Please stay tuned for when data downloads are re-enabled. Thanks.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57272,
      "author_name": "byang1",
      "author_url": "",
      "post_date": "11/04/2014 05:20:54",
      "content": "<p>[quote=Steve Wang;56818]</p>\n<p>...</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>Do you mean C1 and C17 to C24 are all categorical variables ?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57273,
      "author_name": "stevewang0",
      "author_url": "",
      "post_date": "11/04/2014 05:51:44",
      "content": "<p>[quote=B Yang;57272]</p>\n<p>[quote=Steve Wang;56818]</p>\n<p>...</p>\n<p>others are kept private due to business issues.</p>\n<p>besides, all integer features are categorical variables, they are all IDs and have no numerical meaning</p>\n<p>[/quote]</p>\n<p>Do you mean C1 and C17 to C24 are all categorical variables ?</p>\n<p>[/quote]</p>\n<p>yes</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57366,
      "author_name": "payam1",
      "author_url": "",
      "post_date": "11/06/2014 00:12:30",
      "content": "<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57373,
      "author_name": "stevewang0",
      "author_url": "",
      "post_date": "11/06/2014 03:16:54",
      "content": "<p>[quote=Payam;57366]</p>\n<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>\n<p>[/quote]</p>\n\n<p>Hello, we have informed our Kaggle admin to do the editing. Thanks!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57381,
      "author_name": "",
      "author_url": "",
      "post_date": "11/06/2014 05:06:21",
      "content": "<p>[quote=Payam;57366]</p>\n<p>@Steve Wang. It would be nice if these valuable information about the data type, i.e., C1 and C17 to C24 being categorical variables, were&nbsp;made available in the Data description page. Readers should not dig into discussion forums to find out these informations.</p>\n<p>In general, the Data description page is very pure. At least, data types should be released for the challengers. Unless you want only expert data scientists to get involved in this challenge and let beginners to struggle.&nbsp;</p>\n<p>[/quote]</p>\n<p>It does not matter if someone is expert or beginner; missing data types causes unnecessary problems for everyone. It simply make no sense to guess these meta data in anyway. Who knows if any of the assumptions still holds for test data or not?</p>\n<p>I did suspect that some of the variables is ordinal, and that's a relevant piece of information.</p>\n<p>By the way, could the admin explain a little bit more about `device_make`, `device_model` and `device_type`? I can't really understand what do they mean exactly.</p>\n<p>And is there any&nbsp;hierarchical information in the features (like app_category&nbsp;&gt; app_domain &gt; app_id)? That could be relevant too, like once&nbsp;Y. Koren used taxonomy information of predicting music rating in&nbsp;KDDCup 2011.</p>\n<p>Ref:</p>\n<p>http://www.eng.tau.ac.il/~noamk/papers/DKK11.pdf</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57389,
      "author_name": "liamhuber",
      "author_url": "",
      "post_date": "11/06/2014 07:15:41",
      "content": "<p>[quote=byronyi;57381]</p>\n<p>By the way, could the admin explain a little bit more about `device_make`, `device_model` and `device_type`? I can't really understand what do they mean exactly.</p>\n<p>[/quote]</p>\n<p>Make and model--if admins follow similar language to automobiles--is manufacturer and specific device. E.g. (make)(model) = (honda)(civic), or in our case (apple)(iphone6).</p>\n<p>I'd like some clarification on &quot;device_type&quot; too... it's smart phone/tablet/pc/etc maybe?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 57394,
      "author_name": "shuimuxixiha",
      "author_url": "",
      "post_date": "11/06/2014 09:39:50",
      "content": "<p>just say Hi!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 58294,
      "author_name": "poojaphutela",
      "author_url": "",
      "post_date": "11/18/2014 10:04:02",
      "content": "<p>Hi</p>\n<p>Could anyone let me know when Data downloads would be available?</p>\n<p>I am getting below message when click on Data Section.</p>\n<p>&quot;Update - Data downloads are temporarily disabled. Please see&nbsp;the forums&nbsp;for updates.&quot;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 58299,
      "author_name": "poojaphutela",
      "author_url": "",
      "post_date": "11/18/2014 10:39:49",
      "content": "<p>Thanks for the updates !</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 58682,
      "author_name": "miquelmartin",
      "author_url": "",
      "post_date": "11/22/2014 14:13:52",
      "content": "<p>Hello all,</p>\n<p>I've only been around since the new data set was released. In the old posts on this thread, there seems to be more info on what the different fields are. Concretely, <strong>device_os,</strong> <strong>device_make,</strong> and <strong>device_geo_country&nbsp;</strong>are gone in the new dataset description. Is this on purpose? Are they still there, behind C1 or C14-C21?</p>\n<p>Thanks!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 58853,
      "author_name": "changyangfan",
      "author_url": "",
      "post_date": "11/25/2014 05:59:02",
      "content": "<p>why the data have no ad id?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59075,
      "author_name": "changyangfan",
      "author_url": "",
      "post_date": "11/27/2014 08:30:22",
      "content": "<p>Data in order to predict whether the advertisement is to click.However, we have no advertising information. Could you tell me whether the first field data are wrong.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59077,
      "author_name": "changyangfan",
      "author_url": "",
      "post_date": "11/27/2014 08:35:02",
      "content": "<p>I didn't find any 2 identical ID advertising</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 59309,
      "author_name": "sandeepkumar",
      "author_url": "",
      "post_date": "12/01/2014 06:21:21",
      "content": "<p>Hi Steve,</p>\n<p>Could you please tell me whether we need to include C1 and C15 - C24 in the analysis as they are just id's and&nbsp;did not give any description for these id's.</p>\n<p>Please clarify.</p>\n\n<p>Thanks,</p>\n<p>Sandeep</p>",
      "votes": null,
      "replies": [
        {
          "id": 236361,
          "author_name": "hariramn147",
          "author_url": "",
          "post_date": "10/27/2017 05:44:57",
          "content": "<p>Hi Sandeep, If you happened to know the answer for this kindly share </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 60132,
      "author_name": "lemoner",
      "author_url": "",
      "post_date": "12/13/2014 04:44:10",
      "content": "<p>hi all,</p>\n<p>what is the meaning of this feild ?</p>\n<p>'banner_pos'</p>\n<p>thank !</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "56703": "",
    "56704": "",
    "56805": "",
    "56818": "",
    "56820": "",
    "56823": "",
    "56825": "",
    "56833": "",
    "56834": "",
    "56872": "",
    "56873": "",
    "57272": "",
    "57273": "",
    "57366": "",
    "57373": "",
    "57381": "",
    "57389": "",
    "57394": "",
    "58294": "",
    "58299": "",
    "58682": "",
    "58853": "",
    "59075": "",
    "59077": "",
    "59309": "",
    "60132": "",
    "236361": "Hi Sandeep, If you happened to know the answer for this kindly share",
    "969717": "stevewang0 , can I use this dataset for academic publication? thanks.",
    "1877012": "Thank you for sharing",
    "2071829": "Thank you for sharing :)",
    "2188128": "Hello,\nI wish to download datasets for CTR prediction for a research work. can you guide me here",
    "2563345": "Thanks for the data that makes me a machine learning player."
  },
  "source": "meta"
}