{
  "fetched_at": "2026-10-20T01:53:17Z",
  "source_url": "https://www.kaggle.com/competitions/amazon-qlearn/data",
  "pages": [
    {
      "name": "data-description",
      "content": "<h3><strong>Class Descriptions</strong></h3>\r\n<p>There are six class codes (0 to 5) corresponding to respective question types as follows:</p>\r\n<p>0 - <strong>Abbreviation</strong></p>\r\n<p>1 - <strong>Human</strong></p>\r\n<p>2 - <strong>Location</strong></p>\r\n<p>3 - <strong>Description</strong></p>\r\n<p>4 - <strong>Entity</strong></p>\r\n<p>5 - <strong>Numeric</strong></p>\r\n<h2><strong>File descriptions</strong></h2>\r\n<ul>\r\n<li><strong>train_questions.txt</strong> - This has two columns, viz., Question Id and Question text. It has 3000 data points that you shall use for training.</li>\r\n<li><strong> train_features.txt</strong> - This is the bag of words representation for the training data with 3000 data points. The first column is the question Id. The remaining columns is the sparse representation of the question features. For eg. a hypothetical row 6, 1014:1 2034:3 represents question Id 6 that has 1 occurrence of word 1014 and 3 occurrences of word 2034.</li>\r\n<li><strong>train_labels.csv</strong> - This has two columns, viz., Question Id and Question class. It has 3000 rows. For eg a row 10,5 means that question Id 10 has a class of 5 (Numeric).</li>\r\n<li><strong>test_questions.txt</strong> - This has two columns, viz., Question Id and Question text. It has 8639 data points for which you shall have to predict the correct classes.</li>\r\n<li><strong>test_features.txt</strong> - This is the Bag of Words representation of above questions. Format similar to train_features.txt but with 8639 data points.</li>\r\n<li><strong>sample_submission_file.csv </strong>- This is a sample submission file for your reference. Please note that the format should be strictly adhered to, otherwise the evaluation shall be incorrect. <strong>There should be no extra information or punctuation marks (not even spaces).</strong></li>\r\n</ul>\r\n<p>We are thankful to Quora, Stanford (SQuAD) and Li and Roth for providing us a sample of the dataset.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<p>Automatic question answering is an emerging field in artificial intelligence. IBM Watson was one system which famously defeated humans in a quiz contest called Jeopardy in 2011. This challenge pertains to the first phase of QA, where you are asked to develop an algorithm which can understand questions automatically. This is a multiclass classification problem, where you will have to design and train a classifier that can categorize a question into one of 6 categories (<strong>human, location, description, entity, abbreviation or numeric</strong>) depending upon what the question pertains to.</p>\r\n<h3><strong>UPDATE:</strong></h3>\r\n<p>1. The limit on the number of submissions per day has been changed to 5.</p>\r\n<p>2. We have a few dummy submissions on the leaderboard. These are from the users \"RishabhDeshmukh\", \"PrinceJain\" and \"OpenDay17\", and were done for some internal checks. These will not be used in the final evaluation and you can safely ignore them.</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<h4><strong>Winners shall be decided based on following aspects:</strong></h4>\r\n<h5><strong>1. Test Set Accuracy</strong></h5>\r\n<p>Please note that the public leaderboard is generated only through a sample of test data and is only for your own evaluation and validation, this data shall not be considered for final evaluation.</p>\r\n<p>Evaluation shall be based upon the accuracy of remaining test data.</p>\r\n<h5><strong>2. Report (Mandatory to be submitted by 12/03/2017, 11:00pm IST)</strong></h5>\r\n<p>At the end of the contest you shall have to provide us a report <strong>not exceeding two pages or 5 slides </strong>briefly describing your approach to build the final classifier. This component carries significant weight in the final evaluation.</p>\r\n<p><strong>3. Code + README file (Mandatory to be submitted by 12/03/2017, 11:00pm IST)</strong></p>\r\n<p>At the end of the contest you shall have to provide us with the code along with README file to enable us to verify your results.</p>\r\n<h3><strong>Email Submission Format</strong></h3>\r\n<p>Send in your documentation and code to qlearnopenday17@gmail.com.</p>\r\n<p>Subject:: &lt;Kaggle_Username&gt; Final Report</p>\r\n<p>File: &lt;kaggle_username&gt;_report.pdf/ppt</p>\r\n<p>Code + README File: &lt;kagle_username&gt;_code.zip</p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "<ol>\r\n<li>Online registration at the Open Day website is mandatory.</li>\r\n<li>This is an individual event.</li>\r\n<li>This event is open to both school and college students.</li>\r\n<li>The participants should have basic programming skills and elementary knowledge of machine learning. We will also provide some tutorials for you to get started on the challenge.</li>\r\n<li>This contest is open to everyone (not just students), however only students who register on open day website shall be eligible for prizes.</li>\r\n<li>You are not allowed to use any external data for training. Your model should be built only from the training data provided.</li>\r\n<li>The decision of the judges will be final.</li>\r\n</ol>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Resources",
      "content": "<p>1. <a href=\"http://events.csa.iisc.ernet.in/opendays2013/twitminer/getting-started.html\" target=\"_blank\">A brief tutorial on ML</a></p>\r\n<p>2. <a href=\"http://disp.ee.ntu.edu.tw/~pujols/Machine%20Learning%20Tutorial.pdf\" target=\"_blank\">A detailed tutorial on ML</a></p>\r\n<p>3. <a href=\"http://www.vikparuchuri.com/blog/natural-language-processing-tutorial/\" target=\"_blank\">A tutorial on NLP</a></p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Frequently Asked Questions",
      "content": "<p>1. Who can participate in the contest?<br />The contest is open to everyone. However, participants who are not students would not be eligible for prizes.</p>\r\n<p><br />2. What programming languages can be used?<br />Any programming language of your choice can be used.</p>\r\n<p><br />3. For the train data and the test data, will we be given only question text or question features as well?</p>\r\n<p>You will be provided with both the raw question text and feature vectors for each question. You do not necessarily have to work on feature vectors provided by us. You can use any feature extraction algorithm of your choice on the question text provided to you.</p>\r\n<p><br />4. Can you suggest some links for learning Machine learning? It will be very helpful.<br />Please visit the event website.</p>\r\n<p><br />5. Do I need to come to IISc to participate in the contest?<br />No, this is an online event; you only need to register online.</p>\r\n<p><br />6. Can a team have more than one member?<br />No, this is an individual event.</p>\r\n<p><br />7. Can we use codes that are available online?<br />Yes, you can use any resource that is publicly available. Please cite them properly when you send us a report on your approach.</p>\r\n<p><br />8. Can we use training data from an external source in addition to the one provided?<br />No, to ensure fairness, this is strictly disallowed and will lead to disqualification.</p>\r\n<p><br />9. Can use algorithms already published in papers?<br />Yes, you can use algorithms published in papers. Please cite them properly when you send us a report on your approach.</p>\r\n<p><br />10. Can we use GUI-based tools like Weka, RapidMiner etc? If so, how can we submit the code?<br />You can of course use these tools. In that case you will not be able to submit the code obviously, but you should write in your report the sequence of steps you took on the GUI to arrive at your result. As there are some points for the code also, you will lose out on those points. Therefore, even if you are using weka, try to call the libraries of Weka using code rather than using GUI.</p>\r\n<p><br />11. Can we use tools for part-of-speech tagging, sentiment analysis etc?<br />You can use any tool you want. However these should be documented properly when you send us a report on your approach.</p>\r\n<p><br />12. Can we use information other than the question text and the label while learning a model?<br />You can use external data sources in the training phase (while learning a model); for example, you can use a publicly available list of stop words or an external data source to create features that might be useful for this task but you should not use additional training examples. However, you have to make predictions on the test data using only the learned model; you are not allowed to use an external means of making predictions on the validation / test set such as crawling the web or manually labelling examples, this will lead to disqualification.</p>\r\n<p><br />13. Are we allowed to use a publicly available list of stop words?<br />Yes, but do document this in the report that you would send us at the end.</p>\r\n<p><br />14. What do the scores in the leaderboard tell<br />The leaderboard contains the best validation set accuracy (fraction of validation questions classified correctly) attained by each participant (sorted in descending order). This does not include test set accuracy and is only for your own evaluation. The final evaluation will be done on a test set.</p>\r\n<p>For further queries  visit <a href=\"http://events.csa.iisc.ac.in/openday2017/qlearn_faq.html\" target=\"_blank\">FAQs</a></p>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}