This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Use the algorithm below for the update function for all nodes, where j is the node number, and i is the row of the vector. 6. 7. Similarity – they both attempt to minimize the sum of squared error when learning from data Difference – neural networks can predict nonlinear models 8. ChiSquared= (126.4)^2/6.4+(49.6)^2/9.6+(24)^2/4+(86)^2/6+(47.6)^2/7.6+(1511.4)^2/11.4 9. x coordinate (percent false positives) = FP/ All Neg = 16/(65+16) = 19.75% y coordinate (percent true positives) = TP/All Pos = 82/(82+22)=78.85% Amount made =82 TP * 100 dollar reward = $8200 Cost of campaign = 15 dollars * 185 total people = $2775 Net profit = 8200 – 2775 = $5425...
View
Full Document
 Fall '09
 Merz
 Statistics, Addition, Regression Analysis, statistically significant improvement, percent true positives, percent false positives

Click to edit the document details