Scribd
Upload
6K views

Machine Learning Axiom

This document provides information about various machine learning algorithms and concepts through a series of multiple choice questions. Supervised learning techniques like decision trees and naive Bayes apply conditional probability to dependent and independent variables. Unsupervised learning techniques like clustering group data without labels, and kernel methods can be used for both supervised and unsupervised problems. Gradient descent is useful for nonlinear regression as there is no closed form solution. SVM and logistic regression are commonly used for classification.

Uploaded by

Mahesh VP
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
6K views

Machine Learning Axiom

This document provides information about various machine learning algorithms and concepts through a series of multiple choice questions. Supervised learning techniques like decision trees and naive Bayes apply conditional probability to dependent and independent variables. Unsupervised learning techniques like clustering group data without labels, and kernel methods can be used for both supervised and unsupervised problems. Gradient descent is useful for nonlinear regression as there is no closed form solution. SVM and logistic regression are commonly used for classification.

Uploaded by

Mahesh VP
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Machine Learning Axiom

------------------------------

1. Which of the learning methodology applies conditional probability of all the variables with respective
the dependent variable? - supervised learning
2. Can you make quick guess where Decision tree will fall into ___ - Supervised Learning
3. Do you think heuristic for rule learning and heuristics for decision trees are both same ?-False
4. What is the benefit of Naïve Bayes ? - requires less training data

5. What is the advantage of using an iterative algorithm like gradient descent ? (select the best)- For
Nonlinear regression problems, there is no closed form solution
6. For which one of these relationships could we use a regression analysis? Choose the correct one -
Relation between height and weight (both quantitative)
7. Does Logistic regression check for the linear relationship between dependent and independent
variables ? - False
8. Which helps SVM to implement the algorithm in high dimensional space? - Kernel
9. Kernel methods can be used for supervised and unsupervised problems --True
10. Perceptron is - a single layer feed-forward neural network
11. While running the same algorithm multiple times, which algorithm produces same results? -
Hierarchical clustering

12. The model which is widely used for the classification is - Segmentation x,
13. Which of the following is not example of Clustering? - RFM Analysis
14. SVM will not perform well with large data set because - Lot of noise in data x,
15. SVM will not perform well with data with more noise because - target classes could overlap
16. Most famous technique used in Text mining is - Naive Bayes
17. If you have a basket of different fruit varieties with some prior information on size, color, shape of each
and every fruit . Which learning methodology is best applicable - Supervised Learning

18. In Kernel trick method, We do not need the coordinates of the data in the feature space -True

19. Which technique implicitly defines the class of possible patterns by introducing a notion

20. of similarity between data? Kernel

21. Which model helps SVM to implement the algorithm in high dimensional space? - classification--wrong,
Multi-Linear Regression x,

22. The main problem with using single regression line - merging of groups x,

23. If the outcome is binary(0/1), which model to be applied? -Linear regression x,Classification

24. Which of them, best represents the property of Kernel? -Modularity


25. Objective of unsupervised data covers all these aspect except - prepare the training data set

26. Which of the learning methodology applies conditional probability of all the variables with respective
the dependent variable? -supervised learning

27. Correlation and regression are concerned with the relationship between _________ - 2 quantitative
variable

28. The main difficulty with using a regression line to analyze these data is _______ - presence of 1 or more
outliers

29. Disadvantage of Neural network according to your purview is : takes a long time to be trained

30. Which methodology works with clear margins of separation points? - Support Vector Machine

31. The model which is widely used for the classification is - segmentation x,Logistic Regression

32. SVM uses which method for pattern analysis in High dimensional space? Kernel

33. Which type of the clustering could handle Big Data? -K-means

34. In a scenario, where the statistical model describes random error or noise instead of underlying
relationship, what happens - Overfitting

35. If the outcome is continuous, which model to be applied? -Linear regression

36. Does Logistic regression check for the linear relationship between dependent and

37. independent variables ? F

38. For which one of these relationships could we use a regression analysis? Chose the correct one -
Relation ship between weith and height

39. What are different types of Supervised learning-regression and classification

40. One has to run through ALL the samples in your training set to do a single update for a parameter in a
particular iteration. This is applicable for -Gradient descent

41. The correlation between two variables is given by r = 0.0. . This means -There is a perfect positive
relationship between the two variables--wrong, The best straight line through the data is horizontal.

42. Effect of outlier on the correlation coefficient -- decrease the correlation--wrong ,An outlier might
either decrease or increase a correlation coefficient, depending on where it is in relation to the other
points
43. Which clustering technique requires prior knowledge of the number of clusters required -- K Means
clustering

You might also like