Machine Learning models are useful in solving a number of business problems, but there is always a trade-off between accuracy and interpretability of models. Businesses tend to use easily interpretable algorithms which come with lower accuracy. In this article, the author attempts to resolve this issue and shows how powerful black-box algorithms can be used for predictions. He has taken the help of one such algorithm called LIME (Locally Interpretable Model-Agnostic Explanations) that can effectively explain the predictions of any regressor by approximating it locally with an interpretable model. Read more at: https://www.analyticsvidhya.com/blog/2017/06/building-trust-in-machine-learning-models/