this is for holding javascript data
JP Breuer added Now_using_the_actual_information__.tex
about 8 years ago
Commit id: def369dbc68bb95eefb581ce8d9f408a5d4a1916
deletions | additions
diff --git a/Now_using_the_actual_information__.tex b/Now_using_the_actual_information__.tex
new file mode 100644
index 0000000..7f2d0e7
--- /dev/null
+++ b/Now_using_the_actual_information__.tex
...
Now using the actual information that was removed for each observation, a confusion matrix was created from using the Gaussian Mixture Model. As can be seen in Figure 8, there was a relatively high error rate despite the optimization parameters that were chosen for the clustering, and based on the output, it is possible to see how each class of cars was misclustered. This misclustering could come from the fact that despite the vast differences in engineering regarding the mechanics of cars that are produced in different world regions, ultimately these differences are not enough in the principal components to be able to by themselves predict or make any conclusive remarks or observations. For this dataset, using different clustering methods like the K-means method, or another distance metric, will give wildly different results than the algorithms that were chosen, therefore further analysis must be done.
