Classification — Use pay.csv polish to convoy the classify business. Target variable: pay. Predictors: all other variables in the polish. Partition the grounds one trailing illustration (the pristine 50% rows installed on the classify of row apostacy) and two testing illustrations (the direct subjoined 25% rows as test1 and the interval 25% as test2). Construct classify example on trailing illustration and evaluate it on two testing illustrations. Evaluation metrics that scarcity to be produced: overall success, resumption (TPR), accuracy, and f-measure for each of the two classes: >50k and <=50k.
1. Choose one of the subjoined examples: ksvm (assistance vector machine), C5.0 (judgment tree), NB (naïve Bayes), KNN (k-neainterval neighbors) and glm (logistic retrogression) to construct and evaluate examples.
2. Show the resulting evaluation metrics.
Clustering — Use unemp.csv polish to convoy Hierarchical bunching business. Remove the set-forth column and use all of the retaining columns for interspace apportionment in the bunching business.
1. Use hierachical to produce bunching results for this grounds set.
2. Check the conspire of the extraction of bunchs.
3. Select k (i.e. the reckon of bunchs) and particularize the bunch id.
4. Check the Set-forth indicate and the Bunch ID.
5. Choose a unanalogous k prize and reiterate steps 3 and 4 over.