26. Machine learning tasks
• Predictive Tasks
Use some variable to predict unknown or future values of other
variables
It determine what might happen in future
• Descriptive Tasks
Find human-interpretable patters n that describe the data.
It describe ‘what’ happened in past
27. Techniques
• Classification(Prediction)
• Clustering(Descriptive):grouping
• Association rule Discovery(Descriptive):Produce dependency
rules
• Sequential Pattern Discovery(Descriptive): Highlights patterns
on temporal sequences
• Regression(Prediction):Linear and multiple linear regression
• Deviation Detection(Descriptive): Outliers & exceptional data
28.
29.
30.
31.
32.
33.
34.
35.
36. Performance Metrics
Accuracy
Accuracy =
𝑇𝑃 + 𝑇𝑁
𝑇𝑃 + 𝑇𝑁 + 𝐹𝑃 + 𝐹𝑁
Precision
Precision =
𝑇𝑃
𝑇𝑃 + 𝐹𝑃
Recall or Sensitivity
TPR =
TP
TP + FN
F-Score
F1 − score = 2 ∗
precision ∗ TPR
precision + TPR
Receiver Operating Characteristic Curve The receiver
operating characteristic curve (ROC curve) is a graph that displays how well a
classification model performs across all categorization levels. An ROC curve plots TPR
vs. FPR at different classification threshold
𝑇𝑃𝑅 =
𝑇𝑃
𝑇𝑃 + 𝐹𝑁
𝐹𝑃𝑅 =
𝐹𝑃
𝐹𝑃 + 𝑇𝑁
Actual Class
Positive Negative
Predicted
Class
Positive TP FP
Negative FN TN
• ConfusionMatrix