{
  "id": 23044,
  "title": "Some basic statistics on features",
  "url": "/competitions/bosch-production-line-performance/discussion/23044",
  "author_name": "",
  "post_date": "2016-08-23T05:38:13.223Z",
  "votes": 11,
  "comment_count": 2,
  "views": 1360,
  "content": "<p>Dataset contains an extremely large number of anonymized features. Dataset size is quite large 14GB. It contains numerical, categorical and date features. Date feature indicates the test time of each feature. </p>\n\n<p>&#8216;Id&#8217; represents the label of component.\n&#8220;Response&#8221; indicates whether this component is faulty. 1 is faulty, 0 is normal.</p>\n\n<p>The goal is to determine if a component is faulty based on [Test time of each feature, NO. of Line, No. of station, No. of feature, values of each Features]</p>\n\n<p>This is an imbalanced dataset&#65306;</p>\n\n<p>Several reasons may explain why the dataset contains enamours amount of NAN.\n&#8226;   Each component has limited features and only be tested on its own features.\n&#8226;   Some features are not tested.</p>\n\n<p>Data structure: \nThere are 4 assembly lines [0,1,2,3]\nEach line has some stations: Which station is on which line? \nEach station is used to test some features: How many features are tested on each station?\nThere are 51 stations </p>",
  "messages": [
    {
      "id": "132162",
      "postDate": "08/23/2016 05:38:13",
      "content": "<p>Dataset contains an extremely large number of anonymized features. Dataset size is quite large 14GB. It contains numerical, categorical and date features. Date feature indicates the test time of each feature. </p>\n\n<p>&#8216;Id&#8217; represents the label of component.\n&#8220;Response&#8221; indicates whether this component is faulty. 1 is faulty, 0 is normal.</p>\n\n<p>The goal is to determine if a component is faulty based on [Test time of each feature, NO. of Line, No. of station, No. of feature, values of each Features]</p>\n\n<p>This is an imbalanced dataset&#65306;</p>\n\n<p>Several reasons may explain why the dataset contains enamours amount of NAN.\n&#8226;   Each component has limited features and only be tested on its own features.\n&#8226;   Some features are not tested.</p>\n\n<p>Data structure: \nThere are 4 assembly lines [0,1,2,3]\nEach line has some stations: Which station is on which line? \nEach station is used to test some features: How many features are tested on each station?\nThere are 51 stations </p>",
      "rawMarkdown": "Dataset contains an extremely large number of anonymized features. Dataset size is quite large 14GB. It contains numerical, categorical and date features. Date feature indicates the test time of each feature. \r\n\r\n‘Id’ represents the label of component.\r\n“Response” indicates whether this component is faulty. 1 is faulty, 0 is normal.\r\n\r\nThe goal is to determine if a component is faulty based on [Test time of each feature, NO. of Line, No. of station, No. of feature, values of each Features]\r\n\r\nThis is an imbalanced dataset：\r\n\r\nSeveral reasons may explain why the dataset contains enamours amount of NAN.\r\n•\tEach component has limited features and only be tested on its own features.\r\n•\tSome features are not tested.\r\n\r\nData structure: \r\nThere are 4 assembly lines [0,1,2,3]\r\nEach line has some stations: Which station is on which line? \r\nEach station is used to test some features: How many features are tested on each station?\r\nThere are 51 stations",
      "votes": null
    },
    {
      "id": "132415",
      "postDate": "08/25/2016 10:37:47",
      "content": "<p><img src=\"http://desktop/bosch_results/line_and_feature.png\" alt=\"line_and_station\" title></p>",
      "rawMarkdown": "![line_and_station][1]\r\n\r\n\r\n  [1]: http://Desktop/bosch_results/line_and_feature.png",
      "votes": null
    },
    {
      "id": "132416",
      "postDate": "08/25/2016 10:39:14",
      "content": "",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 132415,
      "author_name": "kaigeyang",
      "author_url": "",
      "post_date": "08/25/2016 10:37:47",
      "content": "<p><img src=\"http://desktop/bosch_results/line_and_feature.png\" alt=\"line_and_station\" title></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 132416,
      "author_name": "kaigeyang",
      "author_url": "",
      "post_date": "08/25/2016 10:39:14",
      "content": "",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "132162": "Dataset contains an extremely large number of anonymized features. Dataset size is quite large 14GB. It contains numerical, categorical and date features. Date feature indicates the test time of each feature. \r\n\r\n‘Id’ represents the label of component.\r\n“Response” indicates whether this component is faulty. 1 is faulty, 0 is normal.\r\n\r\nThe goal is to determine if a component is faulty based on [Test time of each feature, NO. of Line, No. of station, No. of feature, values of each Features]\r\n\r\nThis is an imbalanced dataset：\r\n\r\nSeveral reasons may explain why the dataset contains enamours amount of NAN.\r\n•\tEach component has limited features and only be tested on its own features.\r\n•\tSome features are not tested.\r\n\r\nData structure: \r\nThere are 4 assembly lines [0,1,2,3]\r\nEach line has some stations: Which station is on which line? \r\nEach station is used to test some features: How many features are tested on each station?\r\nThere are 51 stations",
    "132415": "![line_and_station][1]\r\n\r\n\r\n  [1]: http://Desktop/bosch_results/line_and_feature.png",
    "132416": ""
  },
  "source": "meta"
}