WorksheetsSeason 2 #Spaic ML Azure Weekly Quiz
Total questions: 20
Worksheet time: 12mins
What is Classification ?
One of the common task in ML
When the expected predicted output is in the form of discrete values
When the expected predicted output is in the form of continues values
One of the common approaches in ML
The type of Classification are :
Binary classification
Regression to arbitrary value
Multi-class, Multi-label classification
Multi-class Single-label classification
Confusion matrix is a matrix where the predicted vs the actual value are placed to see how well the model performed
True
False
What are the evaluation metrics for Classification?
Accuracy (basically the proportion of correct prediction)
Precision ( the proportion of positive cases that are correctly predicted)
Recall ( the proportion of actual positive cases (T) that are correctly predicted )
R squared (How close the regression line to the true value)
There are many different Classification algorithms such as :
Logistic regression
SVM support vector machine
Linear Regression
Decision Forest Regression
To create a model we set the compute target :
Before creating the model
After the creation of the model
Split data component is used to split the data into 2 majors sets ( the training and the testing sets)
True
False
What do we set in the training model component (edit column)?
The input column used for the training
The target column to predict
All the above choice
We use the score model component :
To evaluate the performance of the model
To predict the result of the testing dataset
All the above choice
What are the evaluation model chart for Classification ?
ROC Curve
Precision Recall curve
Histogram of residual
Predict vs True value
What do you conclude after seeing this visualization?
The training error in first plot is maximum as compare to second and third plot.
The best model for this regression problem is the last (third) plot because it has minimum training error (zero).
The second model is more robust than first and third because it will perform best on unseen data.
The third model is overfitting more as compare to first and second.
All will perform same because we have not seen the testing data.
Is Logistic regression a supervised machine learning algorithm?
True
False
Suppose, You applied a Logistic Regression model on a given data and got a training accuracy X and testing accuracy Y. Now, you want to add a few new features in the same data. Select the option(s) which is/are correct in such a case.
Training accuracy increases
Training accuracy increases or remains the same
Testing accuracy decreases
Testing accuracy increases or remains the same
Fraud Detection, Image Classification, Diagnostic, and Customer Retention are applications in
Unsupervised Learning: Clustering
Supervised Learning: Classification
Reinforcement Learning
Unsupervised Learning: Regression
What is common between overfitting and underfitting?
They both have the word fitting in them
Both are errors because of too much simplicity or complexity
They're both very common problems in beginner coding
Which of the following metrics can be used for evaluating regression models?
i) R Squared
ii) Adjusted R Squared
iii) F Statistics
iv) RMSE / MSE / MAE
ii and iv
i and ii
ii, iii and iv
i, ii, iii and iv
How many coefficients do you need to estimate in a simple linear regression model (One independent variable)?
1
2
3
4
Regression is a case of which type of ML?
Supervised
Unsupervised
Reinforced
Deep learning
Describe the correlation in the graph shown.
Strong Negative
Strong Positive
Weak Negative
No Correlation
In regression analysis, the variable that is being predicted is;
the independent variable
the dependent variable
usually denoted by x
usually denoted by r
