{
  "fetched_at": "2026-10-19T20:37:39Z",
  "source_url": "https://www.kaggle.com/competitions/aist4010-spring2022-a3/data",
  "pages": [
    {
      "name": "data-description",
      "content": "- `edges.txt`: This file contains all the edges in the graph. Every row is an edge. For one row (edge), the two columns are the two nodes that the edge connects.\n- `features.txt`: This file contains all nodes in the graph and their features. Every row is a node. For one row (node), the first column is the node name, and the group of rest column elements is the feature vector for this node.\n- `train_labels.csv`: You are supposed to use nodes and their labels in train data set to train your model for the task. The first column is the node name, and the second is the label.\n- `val_labels.csv`: You are supposed to use nodes and their labels in val data set to validate the classification accuracy. The first column is the node name, and the second is the label.\n- `test_idx.csv`: You are supposed to assign classes for nodes in test data set and submit your result. The first column is the node name, and you need to fill in the class in the second column.\n- `sample-submission.csv`: This is a sample submission file for this competition.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "This is the Kaggle part of Assignment 3.\n\nIn this assignment, you will use basic Graph Convolutional Networks (GCNs) or some advanced GNNs to do **node classification**. In this task, your system will be given a whole graph including nodes with their features, edges between nodes, and labels of some nodes as input. You will have all knowledge for the graph structure while you only get part of label information for graph nodes. Here we have 7 classes. The system should decide which class the rest nodes should belong to.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation metric for this competition is accuracy.\n\n$$\naccuracy = \\frac{Number\\ of\\ correctly\\ classified\\ nodes}{Number\\ of\\ total\\ nodes}\n$$\n\n## Submission Format\n\nThe submission '.csv' file should contain two columns: `id` and `label`.\n\nThe file should contain a header and have the following format:\n```\nid,label\n463825,Class_2\n1121569,Class_1\n1105531,Class_1\n...,...\n```",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": " - Don't cheat!\n - Beat the baseline!\n - Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}