Ethical Storyboarding Activity
Write a one-paragraph story describing a fictional person who was positively
affected by a model trained with these data.
The system that predicts the range of Sarah’s risk can be determined by the prediction system. Based on the predicted risk values the range of risk will be assigned. When it's predicted, then it can be treated accurately.
Write a one-paragraph story describing a fictional person who was negatively
affected by a model trained with these data.
Sarah had other factors of knowing signs of breast cancer, all cancer tests are the potential for diagnosing tumors that will not become life-threatening (overdiagnosis) and the possibility of receiving false-positive test results.
Describe at least two sources of bias the particular model in your story could
have.
The model story could have bias in the fact that Sarah could have had breast cancer in the past making her more prone to receiving a positive test. Also Sarah could be one that never had cancer and could be tested and receive false positive results.
Describe at least one way we could modify the model to mitigate this bias.
In order to mitigate this process we will collect enough data to make sure our model fits the data appropriately, not overfitting the data, to give accurate results in our prediction model.
Describe at least one way we could modify the dataset to mitigate this bias.
One way to mitigate this bias in the dataset could be to make sure we create a function to be sure that there is no missing data that could skew our results creating bias.
Describe at least one way you could modify the context surrounding the model to mitigate this bias.
One way to one way you could modify the context surrounding the model to mitigate this bias could be to make sure we create a function where it be relatable to the person