ACCEPTED DATA CHALLENGE

  • Abel  Marie (BSc Student)
  • Birku Abere (BSc Student)
  • Habtamu Ayenew (MSc Student)

Deep learning IndabaXEthiopia is launching a Data Challenge to provide participants with the opportunity to learn and practice Machine Learning (ML) skills on a real-world problem. This event is intended to provide a venue for young talents (particularly undergraduate and first-year postgraduate students) to demonstrate their ML abilities.

  1. Title:  Predicting the Pregnancy Outcome
  2. Sources: Dabat Research Center which is established by University of Gondar 
  3. Relevant Information: The data is related to predicting the outcome of pregnancy. This dataset was collected from 2014 to 2021, and is organized into three tables: demographic information, location of people, and the pregnancy observation related dataset. The expected outcome is predicting the outcome of pregnant women labeled with the "type" feature.
  4. Number of Instances: 14354 for csv 
  5. Number of Attributes: 27 + output attribute.
  6. Missing Attribute Values: 58985 (represented by 999)
  7. The dataset is divided into two parts: training and testing. Files available for download:
  • csv - contains the type (class label). This is the dataset that you will use to train your model.
  • csv- resembles Train.csv but without the type (output-related) columns. This is the dataset on which you will apply your model to predict pregnancy outcome.
  • You have the option of competing as an individual or as a team. Each team should have no more than three members.
  • Participate in the data challenge by registering on IndabaXEthiopia2022.
  • At the end of the competition, you will be able to submit your best score.
    • If two solutions earn identical scores, the tiebreaker will be the date and time in which the submission was made (the earlier solution will win).
    • Your submission file should look like this:

  Registration ID            type(outcome)   Recall    Precision    F1-Score                               ID_000YI58E           0.97

  • The top three Data Challenge solutions will be presented, and the winning individuals and/or teams will receive a special prize.
  • A brief report detailing the methods, tools employed and the results obtained.
  • The accuracy of your perdition model in predicting pregnancy outcomes for separate test data.

Important Dates