{
  "fetched_at": "2026-10-20T01:27:33Z",
  "source_url": "https://www.kaggle.com/competitions/3c-shared-task-purpose/data",
  "pages": [
    {
      "name": "rules",
      "content": "<p>This task is open to the public. All teams are required to submit the code and description of their work. While making a submission using the \"Submit Predictions\", please include the system description as well. We also encourage participants to write a paper describing their work and submit it to the WOSP 2020 workshop. Promising solutions will be invited to give a presentation at WOSP 2020. Different solutions used to approach this task will be documented  in an overview paper provided by the organizers after the end of the shared task.<b>The final interpretation/decision right belongs to the organizing committee of the WOSP 2020 workshop.</b></p>\n<h2>Team Limits</h2>\n<p>There is no maximum team size.</p>\n\n<h2>Submission Limits</h2>\n<p>The limit on the number of submissions per day is set to 20. All teams are allowed to submit a maximum of 5 runs to the competition for the final evaluation.</p>\n\n<h2>One account per team</h2>\n<p>Team submissions should be made through one kaggle account.</p>\n\n<h2>Team Mergers</h2>\n<p>Team mergers are not allowed in this competition.<p>\n\n<h2>Use of External Data</h2>\n<p>Participants are encouraged to use external datasets (ex: ACL-ARC) compatible to the ACT data, for the purpose of training. Use of external datasets should be mentioned in the submitted paper.</p>\n\n<h2>Competition Timeline</h2>\n<p>Start Date: 11/05/2020</p>\n<p>Entry Deadline: None</p>\n<p>End Date: 22/06/2020 11:59 PM UTC</p>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "data-description",
      "content": "<h2>Dataset for 3C Shared Task</h2>\n<h3>Train Dataset</h3>\n<p>The participants will be provided with a labeled training dataset of 3000 instances annotated using the ACT platform.<p>\n                        <p>The dataset is provided in csv format and contains the following fields:<p>\n                        <ul>\n                            <li>Unique Identifier</li>\n  \t\t\t    <li>COREID of Citing Paper</li>\n                            <li>Citing Paper Title</li>\n                            <li>Citing Paper Author</li>\n                            <li>Cited Paper Title</li>\n                            <li>Cited Paper Author</li>\n  \t\t\t    <li>Citation Context</li>\n  \t\t\t    <li>Citation Class Label</li>\n  \t\t\t</ul>\n<p>Each ciation context in the dataset contains the label, \"#AUTHOR_TAG\", which represents the citation that is being considered. The citation class labels corresponds to values between 0 and 5, where each value represents the following categories:\n\n                         <dl>   \n                              <dd>0  -  BACKGROUND</dd>\n                              <dd>1  -  COMPARES_CONTRASTS</dd>\n                              <dd>2  -  EXTENSION</dd>\n                              <dd>3  -  FUTURE</dd>\n                              <dd>4  -  MOTIVATION</dd>\n                              <dd>5  -  USES</dd>\n  \t\t\t </dl>\n</p>\n\n<h3>Test Dataset</h3>\nThe test dataset, in cvs format, contains 1000 instances, with all fields in the train set, except for the labels. ",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h2>Overview</h2>\n<p>The shared task organized as part of the 8th International Workshop on Mining Scientific Publications, <a href=\"https://wosp.core.ac.uk/jcdl2020/index.html\" target=\"_blank\"><strong>WOSP 2020</strong></a> focuses on classifying citation context in research publications based on their influence and purpose.</p> <p>More details on the shared task can be viewed at the<a href=\"https://wosp.core.ac.uk/jcdl2020/shared_task.html\" target=\"_blank\"> 3C Shared task website</a>.</p>\n<p>This competition is for participation in the subtask A, a multi-class classification problem, with 6 classes; Background, Motivation, Uses, Compares_Contrasts, Future and Extension. To participate in subtask B, please click <a href=\"https://www.kaggle.com/c/3c-shared-task-influence\" target=\"_blank\">here</a>.</p> \n\n<p><b>Task participants are required to</b>:\n<ul>\n           <li>Develop methods addressing the task and submit the results via Kaggle</li>\n           <li>Document and submit their method as a short paper as specified on the <a href=\"https://wosp.core.ac.uk/jcdl2020/index.html#submission\" target=\"_blank\"><strong>WOSP 2020</strong></a> website</li>\n           <li>Provide source code for each method</li>\n\n</p>\n\n<b>Paper and Code Submission</b>\n<p>For paper and code submission, please register using <a href=\"https://easychair.org/conferences/?conf=wosp2020\" target=\"_blank\">EasyChair</a>. More information can be found <a href=\"https://wosp.core.ac.uk/jcdl2020/shared_task.html#submission\" target=\"_blank\">here.</a></p>\n<p>The winning team will be awarded a prize, which will be announced later.</p>\n<b>Kaggle Submission deadline: 22 June 2020</b>\n<b>WOSP Paper Submission deadline: 05 July 2020</b>\n\n\n<h2>References</h2>\n<ol>\n       <li>Pride, David, and Petr Knoth. \"Incidental or influential?-challenges in automatically detecting citation importance using publication full texts.\" International conference on theory and practice of digital Libraries. Springer, Cham, 2017.</li>\n       <li>Teufel, Simone, Advaith Siddharthan, and Dan Tidhar. \"Automatic classification of citation function.\" Proceedings of the 2006 conference on empirical methods in natural language processing. 2006.</li>\n       <li>Zhu, Xiaodan, et al. \"Measuring academic influence: Not all citations are equal.\" Journal of the Association for Information Science and Technology 66.2 (2015): 408-427.</li>\n       <li>Jurgens, David, et al. \"Measuring the evolution of a scientific field through citation frames.\" Transactions of the Association for Computational Linguistics 6 (2018): 391-406.</li>\n       <li>Valenzuela, Marco, Vu Ha, and Oren Etzioni. \"Identifying meaningful citations.\" Workshops at the twenty-ninth AAAI conference on artificial intelligence. 2015.</li>\n</ol>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h2>Evaluation</h2>\n<p>The evaluation will be conducted using the withheld test data containing 1000 instances. The evaluation metric used will be the F1-macro score.</p>\n<p><img src=\"https://wosp.core.ac.uk/jcdl2020/site-assets/images/CodeCogsEqn.gif\" alt=\"\"></p>\n<p>We use F1-macro score in order to give more weight to minority classes and to encourage the task participants to focus on the detection of these minority classes, which are particularly important for advancing the field of research metrics beyond just counting citations. However, we recognize that micro-F might be a better metric if our objective was to build a classifier making the least number of mistakes in production.</p>\n<p>We will use 50% of the testset for evaluation initially and this will appear on the public leaderboard as the competition progresses. The rest of the dataset will be used for the final scoring. This result will be displayed on the private leaderboard, based on which the winners will be decided.</p>\n<h2>Submission Format</h2>\n<p>The submission files need to be in CSV format. Each row in the submission file should contain two columns, with values corresponding to the fields, unique_id, and citation_class_label. The unique_id values are textual numbers with a preceding \"CCT\", followed by a number in the range 1 and 1000 (ex. CCT99). The predicted labels should be a single value between 0 and 5 representing the different classes.\n\nThe file should contain a header and have the following format:</p>\n<pre><code>\nunique_id,citation_class_label\nCCT1,0\nCCT2,3\nCCT3,5\nCCT4,0\nCCT5,2\nCCT6,1\nCCT7,0\nCCT8,4\nCCT9,5\nCCT10,0\n......\nCCT997,1\nCCT998,3\nCCT999,4\nCCT1000,0\n</code></pre>\n\n<p>Please make sure that the values in the unique_id and the citation_class_labels field lie in the range mentioned above. Wrong values in the submission file could result in a failed submission. Also, verify the number of instances in the submissions is equal to 1000 before making submissions.</p> \n<p>Submit your CSV file using the 'Submit Predictions' section.\nYou can check your score from the 'My Submissions' section and choose the one to make it final score by checking the 'Use for Final Score' checkbox.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}