{
  "fetched_at": "2026-10-19T19:02:57Z",
  "source_url": "https://www.kaggle.com/competitions/datascienceatraisa/data",
  "pages": [
    {
      "name": "rules",
      "content": "The deadline of this competition is on the 24th of April. Start soon!\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": "<p>The goal of this competition is to predict the first three years cum of a producing well. There are several attributes that would describe a well like the location, completion, production, and geological features. You should be able to construct the label for the training data set from the production data. </p>\n\n<p>The production data file contains production data for the first three months for wells in the training set. Test set has no production data.</p>\n\n<p>For each well <strong>API</strong>  in the test set, you should be able to predict a single <strong>3_yrs_cum_oil</strong>.\nYou'll need to create the data labels for the training set from the \"production_data_train.csv\". The data label that we need to <strong> predict is the cumulative production of the first three years starting from the peak production </strong>. The file \"production_data_test.csv\" contains the production data starting from the early life of a well till the third month after the peak production. You'll need to do apply the pre-processing we have done in the previous milestone on the production data files.\n</p>\n\n<h2>File descriptions</h2>\n<ul>\n<li><strong>Harmony_data.csv</strong> - file contains a batch of features for both the training and test sets</li>\n<li><strong>IHS_data.csv</strong> - file contains another batch of features for both the training and test sets</li>\n<li><strong>test_APIs.csv</strong> - contains list of APIs you'll use for the test set</li>\n<li><strong>sampleSubmission.csv</strong> - a sample submission file in the correct format</li>\n<li><strong>production_data_train.csv</strong> - production data for the training set</li>\n<li><strong>production_data_test.csv</strong> - production data for the test set</li>\n</ul>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "<h2>Project Overview</h2>\n\n<p>Oil Production forecasting is one of the most important problems in the Oil and Gas industry. Accurately estimating the amount of liquid that could be produced by a certain well along its lifetime is very crucial especially from the economical and business side. There are several factors that contribute into it, starting from the well location, geological factors, drilling technologies used by the operator drilling, neighboring wells and their effect on each others.</p>\n<p>All the wells we are dealing with are located in the United States. For this milestone together with the coming ones, we are going to work on estimating the oil production for a certain well, location or group of wells using the data that we going to supply. Problem will be modeled differently from milestone to another. Ending with the productization of the output models.</p>\n<p>The typical behavior of an oil well production is shown in the figure below. The production rate keeps on increasing for the first few months, until it reaches a peak. Then, it starts to decline. Production months before the peak are conventionally ignored due to their usually unpredictable behavior. After the peak month, an oil well starts in the decline phase, which extends throughout  its remaining lifetime. Throughout this project, we will only consider the decline phase. Any statistics, visualizations, or forecasts will only deal with decline phase, and the pre-peak months will be completely removed in the pre-processing phase.</p>\n\n<h2>Milestone Description</h2>\n<p>In this milestone you are going to use Machine Learning to predict three years cumulative production for a certain well. Given the production curve for a well, you should be able to build the label and make use of given well features to build a ML model that estimates the three years cumulative oil production. Well features could be engineering features, geological features, and location features. Well features will be found separately on two files, Harmony files and IHS files.</p>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "<p>Submissions are evaluated using the <a href=\"https://en.wikipedia.org/wiki/Mean_absolute_error\">mean absolute error</a> between the predicted and actual three years cumulative production. MAE measures the average magnitude of the errors in a set of predictions, without considering their direction.</p>\n<h2> Submission File </h2>\n<p>For each well_API in the test set file, you must predict <strong>three_yrs_cum</strong>. The file should contain a header and have the following format:</p>\n<pre>Id,Predicted<br>49021204210100,994265<br>05123362780000,449806<br>33053082080100,189786<br>... </pre>\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Timeline",
      "content": "<ul><li><strong>April 11, 2019</strong> - Entry deadline. You must accept the competition rules before this date in order to compete.</li>\n\n<li><strong>April 20, 2019</strong> - Team Merger deadline. This is the last day participants may join or merge teams.</li>\n\n<li><strong>April 24, 2019</strong> - Final submission deadline.</li><ul>\n\nAll deadlines are at 11:59 PM Cairo time.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}