January 7, 2021
Honorable Madras High Court said that-“1000 Culprits Can Escape but, One Innocent Should Not be Punished “. More or less legal systems around the world follow this principle.
The above motto shows that accuracy is not always the metric that is been sought out for. If you were to design a model that could assist the court in making decisions, then that model should also incorporate this dictum. The primary goal of your model would be to ensure that no innocent be predicted as guilty.
Further, in case of imbalanced datasets accuracy value can be sometimes misleading. Let me illustrate my point with an example, let’s say you have to build a model that has to predict if an accused is guilty or not based on a given set of evidence. If the training data is unbalanced with overwhelming number of guilty cases (Crime Positive) and only a few innocents (Crime Negative), then your model would incorporate this bias. If the test data you use has 1000 rows in which there are 950 guilty (Crime positive) cases and 50 are innocent (Crime Negative). Now, if your model predicts 990 guilty cases (Crime Positive) and 10 innocent (Crime Negative) cases, then your model horribly fails in its objective. It goes against the established legal edict, it incarcerates 80% of the innocents.
Therefore you need a metric that measures how accurately the innocents are vindicated/incarcerated i.e. identifies true Crime Negative or True Negative rate. Here comes in Specificity, it measures exactly that. Specificity is used to check the reliability of the model.
by : Monis Khan
Quick Summary:
Honorable Madras High Court said that-“1000 Culprits Can Escape but, One Innocent Should Not be Punished “. More or less legal systems around the world follow this principle. The above motto shows that accuracy is not always the metric that is been sought out for. If you were to design a model that could assist […]