{
  "fetched_at": "2026-10-19T14:23:26Z",
  "source_url": "https://www.kaggle.com/competitions/birdso-satellite-ml-2022/data",
  "pages": [
    {
      "name": "data-description",
      "content": "## Files\n\n*   **train.csv** - the training dataset consisting of 1,760 bird call recordings, each classified as belonging to a species of bird\n*   **test.csv** - the test set consisting of 16,626 bird call recording, unclassified\n*   **birdsong_metadata.csv** - supplemental information specifying classified species and other relevant classification data regarding each bird call recording\n\n\n## Task\nDevelop and train a neural network on the given train dataset (train.csv) to classify bird calls as belonging to 1 of 88 distinct bird species listed in the metadata (birdsong_metadata.csv), and subsequently test its performance using the provided test dataset (test.csv). \n\nThere are 61 genera of auditory spectral features that can be extracted from the recordings in the datasets, and each species of bird can count on 20 data points. For additional statistics regarding the datasets, please refer to [https://www.kaggle.com/fleanend/birds-songs-numeric-dataset](url) at which the datasets were compiled.\n\nYour submission.csv file should be formatted as id,species for each test set case.\n\n## Submissions\nIn the Code tab below the competition header above, select the \"New Notebook\" button. This should navigate you to a blank local API, called a Kaggle Kernel, in which you will be writing your code. After you have finished programming, select the \"Save Version\" button in the upper right corner. In the pop-up window, select the option \"Save & Run All (Commit)\" from the dropdown menu before confirming \"Save\". To submit your Kernel's generated .csv submission file, navigate back to Code tab on the competition platform, and select the \"Your Work\" tab. Select the first/most recent notebook, and select the \"Data\" tab at the top of the screen under the notebook name. On the right below \"Output\", select the \"submission.csv\" file and submit to the competition.\n\nTo submit your Kaggle Kernel/notebook to event supervisors at machine.learning@birdso.org, navigate to your notebook's editing page and select \"Download Notebook\" from the File dropdown at the top. Upload your notebook to an email addressed  to machine.learning@birdso.org with your team name and number for evaluation.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "## Welcome!\nThis is the page for the competition task of the BirdSO Machine Learning Event!\n\n## Overview\nThere are many distinct features that can distinguish between bird calls, many of which are hard to track with the untrained ear. However, a neural network is very well capable of being able to conduct analysis on up to 61 \"genera of spectral features\" characteristic of 88 different bird species native to the United Kingdom. \n\n\nYour task is to develop a model using machine learning that can accurately associate a given bird call with a bird species, specified in the Data tab above.\n\n\n## Acknowledgements\n\nWe thank xeno-canto: [www.xeno-canto.org](url) for providing the dataset, and Edoardo Ferrante for compiling the train and test datasets.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "- Please submit original work!\n- If results are hardcoded, competitor will receive a 0 for Task Score. ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The metric used to calculate your score will be Categorization Accuracy.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}