A risk chart presents a cumulative performance view of the model.
The x-axis can be thought of as the days across the dataset, but sorting (left to right) to days from the highest probability of rain tomorrow on the left to the lowest probability of rain tomorrow on the right.
The y-axis is then the performance of the model in predicting whether it will rain tomorrow. It is the percentage of the actual days on which it rains that are predicted by the model as raining tomorrow. Thus, 100% (at the top) covers all days on which it rains. For the top 20% of the days with the highest probability of rain tomorrow (Caseload = 20%), some 54% of the actual days for which it rained are predicted by the model.
The more area under the curve the better the model performance. A perfect model would follow the grey line. The Precision line represents the lift offered by the model, with the lift values on the right hand axis.
Your donation will support ongoing availability and give you access to the PDF version of this book. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Popular open source software includes rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0