A confusion matrix is a table that outlines different predictions and test results and contrasts them with real-world values. Confusion matrices are used in statistics, data mining, machine learning models and other artificial intelligence (AI) applications. A confusion matrix can also be called an error matrix.
Confusion matrices are used to make the in-depth analysis of statistical data faster and the results easier to read through clear data visualization. The tables can help analyze faults in statistics, data mining, forensics and medical tests. A thorough analysis helps users decide what results indicate how errors are made rather than merely assessing performance.
Confusion matrices use a simple format to log predictions. In the rows of a confusion matrix for a machine learning model, the possible predictions are aligned on the right-hand side and the actualities along the top. In the rows underneath the actualities, predictions or results are recorded. Results can include the correct indication of a positive as a true positive or a negative as a true negative, as well as an incorrect positive as a false positive or an incorrect negative as a false negative.
Below is a simple example of a confusion matrix that might be used to outline pregnancy test results:
Color coding matrices provide the ability to quickly read a number of instances in a test. Quick visualization makes it possible to analyze a given model’s performance more deeply than a score and to identify trends that might aid in tweaking parameters.