Decision Tree In R | Decision Tree Algorithm | Data Science Tutorial | Machine Learning |Simplilearn
8 likes3,299 views
The document provides an overview of decision trees, explaining their structure and functionality as algorithms for making decisions. It elaborates on their applications in classification and regression, illustrated with examples like predicting survival on a ship based on various factors. Additionally, the document includes a practical use case in R for survival prediction, demonstrating data handling and classification techniques.
![How does a Decision Tree work?
-[2/9.log₂(2/9) + 2/9.log₂(2/9) + 1/9.log₂(1/9) + 4/9.log₂(4/9)]=
-∑ⁱₓ₌₁p(valueₓ).log₂(p(valueₓ))
Shape Size Label Number
round 10 ball 2
rectangle 6 book 2
rectangle 4 card 1
rectangle 2 blocks 4
total 9](https://image.slidesharecdn.com/decisiontreeinrdecisiontreealgorithmdatasciencetutorialmachinelearningsimplilearn-181011104342/85/Decision-Tree-In-R-Decision-Tree-Algorithm-Data-Science-Tutorial-Machine-Learning-Simplilearn-31-320.jpg)
![How does a Decision Tree work?
-[2/9.log₂(2/9) + 2/9.log₂(2/9) + 1/9.log₂(1/9) + 4/9.log₂(4/9)]= 1.8282
-∑ⁱₓ₌₁p(valueₓ).log₂(p(valueₓ))
Shape Size Label Number
round 10 ball 2
rectangle 6 book 2
rectangle 4 card 1
rectangle 2 blocks 4
total 9](https://image.slidesharecdn.com/decisiontreeinrdecisiontreealgorithmdatasciencetutorialmachinelearningsimplilearn-181011104342/85/Decision-Tree-In-R-Decision-Tree-Algorithm-Data-Science-Tutorial-Machine-Learning-Simplilearn-32-320.jpg)
Decision Tree In R | Decision Tree Algorithm | Data Science Tutorial | Machine Learning |Simplilearn
- 2. What’s in it for you? What is a Decision Tree? What problems can be solved using Decision Trees? How does a Decision Tree work? Use Case: Survival prediction in R
- 3. What is a Decision Tree?
- 4. Left or right? What is a Decision Tree?
- 5. Wait or go? Left or right? What is a Decision Tree?
- 6. Decision Tree is a tree shaped algorithm used to determine a course of action. Each branch of the tree represents a possible decision, occurrence or reaction Wait or go? Left or right? What is a Decision Tree?
- 7. What is a Decision Tree? Shopkeeper
- 8. What is a Decision Tree? I must organize my stall.. Shopkeeper
- 9. What is a Decision Tree? Shopkeeper
- 10. What is a Decision Tree? Is it coloured orange? Is it round? No Yes YesNo Shopkeeper
- 11. What problems can be solved using Decision Tree? Classification: Identifying to which set an object belongs Example: Carrot is orange while broccoli is green
- 12. What problems can be solved using Decision Tree? Classification: Identifying to which set an object belongs Example: Carrot is orange while broccoli is green Regression: Regression problems have continuous or numerical valued output variables Example: Predicting the profits of a company
- 13. How does a Decision Tree work? Terms you must know first…..
- 14. How does a Decision Tree work? NODES Each internal node in a decision tree is a test which splits the objects into different categories Is it coloured orange? Is it round? No Yes YesNo Terms you must know first…..
- 15. How does a Decision Tree work? Is it coloured orange? Is it round? No Yes YesNoThis is a node Terms you must know first….. NODES Each internal node in a decision tree is a test which splits the objects into different categories
- 16. How does a Decision Tree work? Is it coloured orange? Is it round? No Yes YesNo ROOT NODE The node at the top of the decision tree is called the Root node Terms you must know first…..
- 17. How does a Decision Tree work? ROOT NODE The node at the top of the decision tree is called the Root node Is it coloured orange? Is it round? No Yes YesNo This is a root node Terms you must know first…..
- 18. How does a Decision Tree work? LEAF NODE Each external node in a decision tree is called the leaf node. The leaf node is the output Is it coloured orange? Is it round? No Yes YesNo Terms you must know first…..
- 19. How does a Decision Tree work? Is it coloured orange? Is it round? No Yes YesNo This is a leaf node LEAF NODE Each external node in a decision tree is called the leaf node. The leaf node is the output Terms you must know first…..
- 20. How does a Decision Tree work? ENTROPY Entropy is a measure of the messiness of your data collection. The messier or more random your data, higher will be the entropy Terms you must know first…..
- 21. How does a Decision Tree work? This collection has high entropy Terms you must know first….. ENTROPY Entropy is a measure of the messiness of your data collection. The messier or more random your data, higher will be the entropy
- 22. How does a Decision Tree work? This collection has high entropy This collection has low entropy Terms you must know first….. ENTROPY Entropy is a measure of the messiness of your data collection. The messier or more random your data, higher will be the entropy
- 23. How does a Decision Tree work? INFORMATION GAIN Information gain is the decrease obtained in entropy by splitting the data set based on some condition Is it coloured orange? Is it round? No Yes YesNo E1 E2 Terms you must know first…..
- 24. How does a Decision Tree work? INFORMATION GAIN Information gain is the decrease obtained in entropy by splitting the data set based on some condition Is it coloured orange? Is it round? No Yes YesNo E1 E2 E1>E2 INFORMATION GAIN=E1-E2 Terms you must know first…..
- 25. How does a Decision Tree work? Hi, my cupboard is a mess. I must organize my stuff..
- 26. How does a Decision Tree work? Classify the objects based on their attribute set using decision trees
- 27. How does a Decision Tree work? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Let’s look at the attributes for each object
- 28. How does a Decision Tree work? • We split at each level based on certain conditions on the attributes Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 29. How does a Decision Tree work? • We split at each level based on certain conditions on the attributes • Splitting aims at reducing the entropy Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 30. How does a Decision Tree work? • We split at each level based on certain conditions on the attributes • Splitting aims at reducing the entropy -∑ⁱₓ₌₁p(valueₓ).log₂(p(valueₓ)) Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 31. How does a Decision Tree work? -[2/9.log₂(2/9) + 2/9.log₂(2/9) + 1/9.log₂(1/9) + 4/9.log₂(4/9)]= -∑ⁱₓ₌₁p(valueₓ).log₂(p(valueₓ)) Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 32. How does a Decision Tree work? -[2/9.log₂(2/9) + 2/9.log₂(2/9) + 1/9.log₂(1/9) + 4/9.log₂(4/9)]= 1.8282 -∑ⁱₓ₌₁p(valueₓ).log₂(p(valueₓ)) Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 33. How does a Decision Tree work? Now we must find the conditions for our split. Every split must give us the maximum achievable information gain Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9
- 34. How does a Decision Tree work? Our first split will be on shape as that will directly segregate the balls Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 E1 = 1.8282
- 35. How does a Decision Tree work? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle? E2 = 1.3784 E1 = 1.8282
- 36. How does a Decision Tree work? Our second split will be on size as that will directly segregate the books Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle? E2 = 1.3784 E1 = 1.8282
- 37. How does a Decision Tree work? Shape == Rectangle? Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 E3 = 0.716 E2 = 1.3784 E1 = 1.8282
- 38. How does a Decision Tree work? Our third split will once again be on size Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle? E3 = 0.716 E2 = 1.3784 E1 = 1.8282
- 39. How does a Decision Tree work? Size>3? Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle? E3 = 0.716 E2 = 1.3784 E1 = 1.8282 E4 = 0
- 40. How does a Decision Tree work? Size>3? Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle? E2 = 1.3784 E3 = 0.716 E1 = 1.8282 E4 = 0
- 41. How does a Decision Tree work? All our objects are now classified with 100% accuracy Size>3? Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle?
- 42. How does a Decision Tree work? All our objects are now classified with 100% accuracy Size>3? Size>5? Shape Size Label Number round 10 ball 2 rectangle 6 book 2 rectangle 4 card 1 rectangle 2 blocks 4 total 9 Shape == Rectangle?
- 43. Use Case: Survival prediction in R Let’s implement classification of a data set based on Information Gain
- 44. Let’s implement classification of a data set based on Information Gain This is the ID3 algorithm Use Case: Survival prediction in R
- 45. Let’s implement classification of a data set based on Information Gain This is the ID3 algorithm We will be using the RStudio IDE Use Case: Survival prediction in R
- 46. • A ship had 20 lifeboats Use Case: Survival prediction in R
- 47. • A ship had 20 lifeboats • The lifeboats were distributed based on the class, gender and age of the passengers Use Case: Survival prediction in R
- 48. • A ship had 20 lifeboats • The lifeboats were distributed based on the class, gender and age of the passengers • We will develop a model that recognises the relationship between these factors and predicts the survival of a passenger accordingly Use Case: Survival prediction in R
- 49. Use Case: Survival prediction in R We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not
- 50. 1 indicates the person survived the wreck We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not Use Case: Survival prediction in R
- 51. The luxury class of the cabin Use Case: Survival prediction in R We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not
- 52. Numbers of siblings on board Use Case: Survival prediction in R We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not
- 53. Numbers of parents on board Use Case: Survival prediction in R We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not
- 54. Disembark location Use Case: Survival prediction in R We will be using a data set which specifies if a passenger on a ship survived it’s wreck or not
- 55. What is DecisionTree? Problems solved using DecisionTrees How does a decision tree work? Key Takeaways Predicting survivors using R Determining accuracy of prediction Terms to know
- 56. So what’s your next step?
Editor's Notes
- #3: Style - 01
- #4: Style - 01
- #5: Style - 01
- #6: Style - 01
- #7: Style - 01
- #8: Style - 01
- #9: Style - 01
- #10: Style - 01
- #11: Style - 01
- #12: Style - 01
- #13: Style - 01
- #14: Style - 01
- #15: Style - 01
- #16: Style - 01
- #17: Style - 01
- #18: Style - 01
- #19: Style - 01
- #20: Style - 01
- #21: Style - 01
- #22: Style - 01
- #23: Style - 01
- #24: Style - 01
- #25: Style - 01
- #26: Style - 01
- #27: Style - 01
- #28: Style - 01
- #29: Style - 01
- #30: Style - 01
- #31: Style - 01
- #32: Style - 01
- #33: Style - 01
- #34: Style - 01
- #35: Style - 01
- #36: Style - 01
- #37: Style - 01
- #38: Style - 01
- #39: Style - 01
- #40: Style - 01
- #41: Style - 01
- #42: Style - 01
- #43: Style - 01
- #44: Style - 01
- #45: Style - 01
- #46: Style - 01
- #47: Style - 01
- #48: Style - 01
- #49: Style - 01
- #50: Style - 01
- #51: Style - 01
- #52: Style - 01
- #53: Style - 01
- #54: Style - 01
- #55: Style - 01
- #56: Style - 01