Example Graphs
Note that the actual numbers in the following examples are made up. These are meant to give you a sense of what the graphs should look like.
Perceptron
Average misclassification rate during training:
![](images/perceptronimg1.png)
Linearly separable instances:
![](images/perceptronimg2.png)
Non-linearly separable instances:
![](images/perceptronimg3.png)
Backpropagation
For many of these graphs, different styles can work. Bar and line graphs are shown below.
![](images/backpropimg1.png)
Line graphs
![](images/backpropimg2.png)
![](images/backpropimg3.png)
Bar graphs
![](images/backpropimg4.png)
![](images/backpropimg5.png)
![](images/backpropimg6.png)
![](images/backpropimg7.png)
Nearest Neighbor
For many of these graphs, different styles can work. Bar and line graphs are shown below.
![](images/nnimg1.png)
![](images/nnimg2.png)
Clustering
The various SSE vs k-value graphs might look something like this:
![](images/clusterimg1.png)
Silhouette graph example
![](images/clusterimg2.png)
Acknowledgments
Thanks to Dr. Tony Martinez for help in designing this course.