{
  "id": 19568,
  "title": "Some ideas on how the feature extraction is done ",
  "url": "/competitions/yelp-restaurant-photo-classification/discussion/19568",
  "author_name": "",
  "post_date": "2016-03-16T20:04:34.747Z",
  "votes": null,
  "comment_count": 1,
  "views": 686,
  "content": "<p>I see that for this challenge, the data are photos, not plain text. So how is the feature extraction done? How many features are we expected to get from the photos? </p>\n\n<p>I am not asking for exact solutions, I am new to the field, so I am confused. Can someone give me a hint? Any techniques, tools or documentation I can browse? </p>\n\n<p>Thank you.</p>",
  "messages": [
    {
      "id": "111802",
      "postDate": "03/16/2016 20:04:34",
      "content": "<p>I see that for this challenge, the data are photos, not plain text. So how is the feature extraction done? How many features are we expected to get from the photos? </p>\n\n<p>I am not asking for exact solutions, I am new to the field, so I am confused. Can someone give me a hint? Any techniques, tools or documentation I can browse? </p>\n\n<p>Thank you.</p>",
      "rawMarkdown": "I see that for this challenge, the data are photos, not plain text. So how is the feature extraction done? How many features are we expected to get from the photos? \r\n\r\nI am not asking for exact solutions, I am new to the field, so I am confused. Can someone give me a hint? Any techniques, tools or documentation I can browse? \r\n\r\nThank you.",
      "votes": null
    },
    {
      "id": "112451",
      "postDate": "03/21/2016 10:50:50",
      "content": "<p>You can check other threads on this forum, such as <a href=\"https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code\">https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code</a> where others discuss and/or provide solutions for image feature extraction.  Generally speaking, when it comes to obtaining semantic information from images, deep learning approaches have been most successful. However, these can be time consuming depending on the hardware used and particular approach.  For less processing intensive approaches, you can look into approaches, such as HOG feature extraction.  I believe that same thread has someone who's posted a HOG-based solution.</p>\n\n<p>Cheers</p>",
      "rawMarkdown": "You can check other threads on this forum, such as https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code where others discuss and/or provide solutions for image feature extraction.  Generally speaking, when it comes to obtaining semantic information from images, deep learning approaches have been most successful. However, these can be time consuming depending on the hardware used and particular approach.  For less processing intensive approaches, you can look into approaches, such as HOG feature extraction.  I believe that same thread has someone who's posted a HOG-based solution.\r\n\r\nCheers",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 112451,
      "author_name": "bluelight773",
      "author_url": "",
      "post_date": "03/21/2016 10:50:50",
      "content": "<p>You can check other threads on this forum, such as <a href=\"https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code\">https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code</a> where others discuss and/or provide solutions for image feature extraction.  Generally speaking, when it comes to obtaining semantic information from images, deep learning approaches have been most successful. However, these can be time consuming depending on the hardware used and particular approach.  For less processing intensive approaches, you can look into approaches, such as HOG feature extraction.  I believe that same thread has someone who's posted a HOG-based solution.</p>\n\n<p>Cheers</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "111802": "I see that for this challenge, the data are photos, not plain text. So how is the feature extraction done? How many features are we expected to get from the photos? \r\n\r\nI am not asking for exact solutions, I am new to the field, so I am confused. Can someone give me a hint? Any techniques, tools or documentation I can browse? \r\n\r\nThank you.",
    "112451": "You can check other threads on this forum, such as https://www.kaggle.com/c/yelp-restaurant-photo-classification/forums/t/19206/deep-learning-starter-code where others discuss and/or provide solutions for image feature extraction.  Generally speaking, when it comes to obtaining semantic information from images, deep learning approaches have been most successful. However, these can be time consuming depending on the hardware used and particular approach.  For less processing intensive approaches, you can look into approaches, such as HOG feature extraction.  I believe that same thread has someone who's posted a HOG-based solution.\r\n\r\nCheers"
  },
  "source": "meta"
}