{
  "fetched_at": "2026-10-24T16:23:00Z",
  "source_url": "https://www.kaggle.com/competitions/datasciencefall2019hw5/data",
  "pages": [
    {
      "name": "rules",
      "content": "This is a page where you can include rules that participants must accept before joining. You may wish to include rules like:\n\n - Don't cheat!\n - Apply yourself!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "<h2>File descriptions</h2>\n<ul>\n<li><strong>train_graph.json</strong> - the training graph, contains nodes and links</li>\n<li><strong>train_graph_id.npy</strong> - the index of group for every nodes</li>\n<li><strong>train_feats.npy</strong> - the features for every nodes</li>\n<li><strong>train_labels.npy</strong> - the class information for every nodes</li>\n\n<li><strong>test_graph.json</strong> - the testing graph, contains nodes and links</li>\n<li><strong>test_graph_id.npy</strong> - the index of group for every nodes</li>\n<li><strong>test_feats.npy</strong> - the features for every nodes</li>\n<li><strong>submission_sample.csv</strong> - the submission sample for testing data</li>\n\n</ul>\n<h2>Keys for the json files</h2>\n<ul>\n<li><strong>directed</strong> - True, indicating that this is a graph with directed link</li>\n<li><strong>nodes</strong> - indicating the index for every node</li>\n<li><strong>links</strong> - indicating the source and target of link, i.e. {'source':0, 'target':3} means that there is a directed link from node 0 to node 3</li>\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "# Graph Node Classification\nIn this competition, you need to predict the classes of every node. There are 45,236 nodes and 4,585,028 links for training data, and 11,708 nodes and 359,180 links for testing data. \nWe also provide graph id, which indicated the group of every nodes (training data contains 20 groups and testing data contains 4 groups). \nThe dimension of features for nodes is 50.\nThere are 121 classes, and each node can be one-class, multi-class or non-class.\n\n## Dataset Description\nThere are 4 files for training:\n1. **train_graph.json**: graph information, i.e. 'nodes', 'links'\n2. **train_graph_id.npy**: the index  for indicating the group of nodes (1-20)\n3. **train_feats.npy**: the features for every nodes, dimension is 50.\n4. **train_labels.npy**: the classes for every nodes, dimension is 121.\n\nThere are 4 files for training:\n1. **test_graph.json**: graph information, i.e. 'nodes', 'links'\n2. **test_graph_id.npy**: the index  for indicating the group of nodes (21-24)\n3. **test_feats.npy**: the features for every nodes, dimension is 50.\n\n<h2>Acknowledgements</h2>\n<p>Citation would be undated after the competition.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h2>Evaluation Metric</h2>\n<p>The evaluation metric for this competition is **Mean F Score**</a>. The F score, commonly used in information retrieval, measures accuracy using the statistics precision p and recall r. Precision is the ratio of true positives (tp) to all predicted positives (tp + fp). Recall is the ratio of true positives to all actual positives (tp + fn). The F1 score is given by:</p>\n<p>\\[ F1 = 2\\frac{p \\cdot r}{p+r}\\ \\ \\mathrm{where}\\ \\ p = \\frac{tp}{tp+fp},\\ \\ r = \\frac{tp}{tp+fn} \\]</p>\n<p>The mean F1 evaluation metric used in this competition is to compute F1 score for each node first, then take the average</p>\n\n<h2>Submission Format</h2>\n<p><strong>For every author in the dataset</strong>, submission files should contain two columns: Id and Predicted. Predicted should be seperated by a space. </p>\n<p>The file should contain a header and have the following format:</p>\n<pre>Id,Predicted<br />0,1 10 13<br /><span style=\"font-size: 14px; line-height: 1.4em;\">1, <br />2,9 10<br /></span><span style=\"font-size: 14px; line-height: 1.4em;\">3,10<br /></span><span style=\"font-size: 14px; line-height: 1.4em;\">etc.</span></pre>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}