Home Courses Cancer Diagnosis using Medical Records ML problem formulation:Mapping real world to ML problem

ML problem formulation:Mapping real world to ML problem

Instructor: Applied AI Course Duration: 19 mins

This content is restricted. Please Login

Prev

Next

ML problem formulation :Data

ML problem formulation:Train, CV and Test data construction

Case Study: Personalized Cancer Diagnosis.

1.1

Business/Real world problem overview

13 min

1.2

Business objectives and constraints.

11 min

1.3

ML problem formulation :Data

5 min

1.4

ML problem formulation:Mapping real world to ML problem

19 min

1.5

ML problem formulation:Train, CV and Test data construction

4 min

1.6

EDA:Reading data & preprocessing

7 min

1.7

EDA:Distribution of Class-labels.

7 min

1.8

EDA:“Random” Model.

19 min

1.9

Univariate Analysis:Gene feature

34 min

1.10

Univariate Analysis:Variation Feature

19 min

1.11

Univariate Analysis:Text feature

15 min

1.12

Data preparation.

8 min

1.13

Baseline Model: Naive Bayes

23 min

1.14

K-Nearest Neighbors Classification.

9 min

1.15

Logistic Regression with class balancing

10 min

1.16

Logistic Regression without class balancing

4 min

1.17

6 min

1.18

Random-Forest with one-hot encoded features

7 min

1.19

Random-Forest with response-coded features

6 min

1.20

Stacking Classifier

8 min

1.21

Majority Voting classifier.

5 min

1.22

5 min