{
  "id": 469209,
  "title": "Has anyone tried using multimodal fusion?",
  "url": "/competitions/hms-harmful-brain-activity-classification/discussion/469209",
  "author_name": "",
  "post_date": "2024-01-19T15:02:23.179428700Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I have a question,can we use multimodal fusion?<br>\nIt is just an idea,I don't know whether this will work.<br>\nOne modality utilizes EEG data, while another modality employs the spectrogram provided by Kaggle.<br>\nAnd I also find a paper,which only use the raw EEG datas without any preprocessing,and get a good accuracy.</p>",
  "messages": [
    {
      "id": "2609546",
      "postDate": "01/19/2024 15:02:23",
      "content": "<p>I have a question,can we use multimodal fusion?<br>\nIt is just an idea,I don't know whether this will work.<br>\nOne modality utilizes EEG data, while another modality employs the spectrogram provided by Kaggle.<br>\nAnd I also find a paper,which only use the raw EEG datas without any preprocessing,and get a good accuracy.</p>",
      "rawMarkdown": "I have a question,can we use multimodal fusion?\nIt is just an idea,I don't know whether this will work.\nOne modality utilizes EEG data, while another modality employs the spectrogram provided by Kaggle.\nAnd I also find a paper,which only use the raw EEG datas without any preprocessing,and get a good accuracy.",
      "votes": null
    },
    {
      "id": "2610210",
      "postDate": "01/20/2024 02:00:30",
      "content": "<p><a href=\"https://www.kaggle.com/kitsuha\" target=\"_blank\">@kitsuha</a> go through this <a href=\"https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467915#2609612\" target=\"_blank\">Discussion comment</a> - <strong>not multimodal</strong></p>\n<blockquote>\n  <p><strong>LB: 0.37</strong> is using both EEG &amp; Spectrograms (Kaggle spectrogram + egg-&gt;spectrogram) sequences. =&gt; by Chris</p>\n</blockquote>\n<p><a href=\"https://www.kaggle.com/kitsuha\" target=\"_blank\">@kitsuha</a> could you share the paper?</p>",
      "rawMarkdown": "kitsuha go through this [Discussion comment](https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467915#2609612) - **not multimodal**\n\n> **LB: 0.37** is using both EEG & Spectrograms (Kaggle spectrogram + egg->spectrogram) sequences. => by Chris\n\n@kitsuha could you share the paper?",
      "votes": null
    },
    {
      "id": "2610291",
      "postDate": "01/20/2024 03:57:31",
      "content": "<p>Sure!Thank you for sharing the discussion.<br>\nYou can find the paper from <a href=\"https://ieeexplore.ieee.org/abstract/document/9207939\" target=\"_blank\">L. Farsi, S. Siuly, E. Kabir and H. Wang, \"Classification of Alcoholic EEG Signals Using a Deep Learning Method,\" in IEEE Sensors Journal, vol. 21, no. 3, pp. 3552-3560, 1 Feb.1, 2021, doi: 10.1109/JSEN.2020.3026830.</a></p>",
      "rawMarkdown": "Sure!Thank you for sharing the discussion.\nYou can find the paper from [L. Farsi, S. Siuly, E. Kabir and H. Wang, \"Classification of Alcoholic EEG Signals Using a Deep Learning Method,\" in IEEE Sensors Journal, vol. 21, no. 3, pp. 3552-3560, 1 Feb.1, 2021, doi: 10.1109/JSEN.2020.3026830.](https://ieeexplore.ieee.org/abstract/document/9207939)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2610210,
      "author_name": "seshurajup",
      "author_url": "",
      "post_date": "01/20/2024 02:00:30",
      "content": "<p><a href=\"https://www.kaggle.com/kitsuha\" target=\"_blank\">@kitsuha</a> go through this <a href=\"https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467915#2609612\" target=\"_blank\">Discussion comment</a> - <strong>not multimodal</strong></p>\n<blockquote>\n  <p><strong>LB: 0.37</strong> is using both EEG &amp; Spectrograms (Kaggle spectrogram + egg-&gt;spectrogram) sequences. =&gt; by Chris</p>\n</blockquote>\n<p><a href=\"https://www.kaggle.com/kitsuha\" target=\"_blank\">@kitsuha</a> could you share the paper?</p>",
      "votes": null,
      "replies": [
        {
          "id": 2610291,
          "author_name": "kitsuha",
          "author_url": "",
          "post_date": "01/20/2024 03:57:31",
          "content": "<p>Sure!Thank you for sharing the discussion.<br>\nYou can find the paper from <a href=\"https://ieeexplore.ieee.org/abstract/document/9207939\" target=\"_blank\">L. Farsi, S. Siuly, E. Kabir and H. Wang, \"Classification of Alcoholic EEG Signals Using a Deep Learning Method,\" in IEEE Sensors Journal, vol. 21, no. 3, pp. 3552-3560, 1 Feb.1, 2021, doi: 10.1109/JSEN.2020.3026830.</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2609546": "I have a question,can we use multimodal fusion?\nIt is just an idea,I don't know whether this will work.\nOne modality utilizes EEG data, while another modality employs the spectrogram provided by Kaggle.\nAnd I also find a paper,which only use the raw EEG datas without any preprocessing,and get a good accuracy.",
    "2610210": "kitsuha go through this [Discussion comment](https://www.kaggle.com/competitions/hms-harmful-brain-activity-classification/discussion/467915#2609612) - **not multimodal**\n\n> **LB: 0.37** is using both EEG & Spectrograms (Kaggle spectrogram + egg->spectrogram) sequences. => by Chris\n\n@kitsuha could you share the paper?",
    "2610291": "Sure!Thank you for sharing the discussion.\nYou can find the paper from [L. Farsi, S. Siuly, E. Kabir and H. Wang, \"Classification of Alcoholic EEG Signals Using a Deep Learning Method,\" in IEEE Sensors Journal, vol. 21, no. 3, pp. 3552-3560, 1 Feb.1, 2021, doi: 10.1109/JSEN.2020.3026830.](https://ieeexplore.ieee.org/abstract/document/9207939)"
  },
  "source": "meta"
}