{
  "fetched_at": "2026-10-19T07:40:29Z",
  "source_url": "https://www.kaggle.com/competitions/digital-turbine-auction-bid-price-prediction/data",
  "pages": [
    {
      "name": "data-description",
      "content": "This section contains important information about the dataset provided. It describes the files we provide you, their format, and the data columns that will be relevant for this competition.\n\n**Each record represents an auction **\n\n\n## Files\n\n*   **train.csv** - the training set\n*   **test.csv** - the test set\n*   **sample_submission.csv** - a sample submission file in the correct format\n\n## Columns\n\n* `eventTimestamp` - request timestamp, in milliseconds, epoch time\n* `unitDisplayType` - ad slot type (e.g. banner, rewarded, ...)\n* `brandName` - device brand name\n* `bundleId` - external app Id (official bundle Id - taken from the app store - Google Play / Apple iOS Store)\n* `appVersion` - external app version (official version)\n* `correctModelName` - specific model device\n* `countryCode` - 2 digit country code, e.g. US\n* `deviceId` - hashed user Id\n* `osAndVersion` - device OS and device OS version, combined into a single column\n* `connectionType` - device connection type\n* `c1` - an anonymized column\n* `c2` - an anonymized column\n* `c3` - an anonymized column\n* `c4` - an anonymized column\n* `size` - ad size (e.g. 320x50)\n* `mediationProviderVersion` - mediation SDK version\n* `bidFloorPrice` - 2 decimal bid floor price we exposed to DT bidders (e.g. 0.01)\n* `sentPrice` - the bid we send into the auction. 2 decimal places (E.g. 1.23)\n* `has_won` - a flag indicating (1) is that DT Exchange won, 0 otherwise\n* `winBid` - the bid that had won in the auction eventually. **Note: This is the target**\n\n### Notes:\n1. `has_won` and `winBid` are not part of the test set <br>\n2. `winBid` is the label of the bid price prediction task and the goal of the competition <br>\n3. `c1`, `c2`, `c3`, `c4` are anonimyzed columns <br>",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "## Welcome to Digital Turbine's Auction Bid Price Prediction!\n\n[Digital Turbine](https://www.digitalturbine.com/) is the leading independent growth & monetization platform, powered by code installed on devices by global carriers and OEMs.\n\nDigital Turbine provides an Ad Exchange for publishers and advertisers - publishers can grow their businesses and monetize their applications, while advertisers get more users installing their apps. The Ad Exchange connects Demand Side Platforms (DSPs) and Supply Side Platforms (SSPs) through [Real Time Bidding](https://en.wikipedia.org/wiki/Real-time_bidding).\n\n [This image](https://prnt.sc/LvUfS0ukBnmv) helps follow the technical flow of an auction. The steps are as follows:\n\n1. An ad request is sent from a user's device into the mediation platform.\n2. The mediation platform initiates an external auction, and sends a bid request to DT Exchange.\n3. DT Exchange receives that bid request, initiates an internal auction on its side using DT Exchange, and sends bid request(s) to its DSPs.\n4. DT Exchange receives the bid response(s) from its bidders, and chooses the winner (1st price auction).\n5. DT Exchange bids accordingly onto the mediation auction (external auction).\n6. The mediation then sends a \"Win / Loss\" auction resolution event to its participants, with winBid (target in this competition).\n7. Finally, an ad response will be sent back to the device, present to the user, who will be able to interact with the chosen ad.\n\n\n\nAs one of the leading exchanges in the world, we operate on a gargantuan scale: we handle 100 billion requests from hundreds of millions of devices daily. One of the most interesting challenges in such Ad Exchange is the pricing problem. Our Ad Exchange is competing with other exchanges and in order to optimize our performance we need to predict what the other exchanges will bid for the request, and what will be the winning bid for the request.\n\nThis is your place to shine and to compete!\n\nGiven a set of auction attributes (features), as well as a signal which indicates if we (DT) had won that auction or not (has_won), your goal is to predict what bid price would have won us the auction (aka the \"winBid\").\nMeaning - the goal is to predict the winning \"winBid\" on a given auction.\n\nNote: The problem is a regression problem (as we're interested in knowing what was the winning bid price in $ [CPM](https://www.investopedia.com/terms/c/cpm.asp).\n\nThe full dataset description can be found under the \"Data\" section, you should refer to it for additional information.\n\n## THIS IS ALSO A RECRUITMENT COMPETITION FOR DATA SCIENCE POSITIONS AT DIGITAL TURBINE\nWe are opening Data Science positions in Warsaw and Petach Tikva.\nIn case you'd like to apply, please see [this position](https://jobs.lever.co/digitalturbine/c1985c0c-3ac1-44df-b8d4-cc7bc0396bb6).\nParticipants with high quality contributions (e.g. a high score on the LB, an innovative notebook, an insightful forum discussion) can secure an interview and a significant advantage in our recruitment process.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation metric for this competition is Root Mean Square Error ([RMSE](https://en.wikipedia.org/wiki/Root-mean-square_deviation)). RMSE is the standard deviation of the residuals (prediction errors). Residuals are a measure of how far from the regression line data points are; RMSE is a measure of how spread out these residuals are. In other words, it tells you how concentrated the data is around the line of best fit\n(see [here](https://www.statisticshowto.com/probability-and-statistics/regression-analysis/rmse-root-mean-square-error/) for further details).\n\nThis metric was chosen due to the fact that errors are squared before averaged. This gives a relatively high weight to large errors, making RMSE the most useful when large errors are particularly undesirable.\n\n## Submission Format\n\nThe file should contain a header and have the following format:\n\nYou should submit a file with two columns: deviceId and winBid. deviceId is the identifier of each separate user in the dataset. winBid is a numerical value, representing the bid that would have won that auction.\n\nThe file should contain a header and have the following format:\n```\ndeviceId,winBid\n12345abcde,1.051\n```",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "## One account per participant\nYou cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n\n## No private sharing outside teams\nPrivately sharing code or data outside of teams is not permitted. It's okay to share code if made available to all participants on the forums.\n\n## Team Mergers\nTeam mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the Team Merger Deadline. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.\n\n## Team Limits\nThe maximum team size is 5.\n\n## Submission Limits\nYou may submit a maximum of 5 entries per day.\n\nYou may select up to 2 final submissions for judging.\n\n## Competition Timeline\nCompetition Timeline dates (including Entry Deadline, Final Submission Deadline, Start Date, and Team Merger Deadline, as applicable) are reflected on the competition’s Overview > Timeline page.\n\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}