LIME works by perturbing the input data and observing the changes in the model's predictions. It creates a new, simpler model that approximates the complex model locally around a specific instance. This simpler model is typically a linear model or a decision tree, which is much easier to interpret.
