Loading project…
I trained KNN, a decision tree, logistic regression and a passive-aggressive classifier on age and salary to predict a purchase, then scored each one on all 400 rows.
Original tutorial by Aman Kharwal: Comparison of Classification Algorithms (opens in a new tab)
| Output | Value | Cell # |
|---|---|---|
| KNN score | 87.5% | |
| Decision tree score | 98.0% | |
| Logistic regression score | 64.3% | |
| Passive aggressive score | 64.3% |
Cell # counts every cell from the top of the notebook, Markdown included, starting at 1.
360 of the 400 scored rows were training rows, so the decision tree's 0.98 is mostly memory. The SVM and Bernoulli naive Bayes models were created but never trained.
A 400-row social network ads table from the tutorial author's data repository.
27
InteractiveConclusion flipsThe tree's 0.98 implies 33 of 40 right on the unseen rows, about 82.5%. Logistic regression and the passive-aggressive model both scored exactly 257 of 400, which is what predicting 'no purchase' for everyone would score.
Read the audit