Viva prep · Real questions · Student experiences Enroll in Bootcamp

Search viva questions

1_MLP16 MLP Level 1 Aug 7, 2026
  1. 1
    Explain your EDA.
  2. 2
    Pick any one EDA visualization and explain it in detail.
  3. 3
    Explain how any one model works.
  4. 4
    Why does Hyperparameter Tuning improve the model score?
  5. 5
    Explain any one Hyperparameter Tuning parameter.
Level3_29 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    What did you learn from this project/course?
  3. 3
    What is Machine Learning?
  4. 4
    Difference between Machine Learning and Data Science.
  5. 5
    Difference between Machine Learning and Deep Learning.
  6. 6
    What are the types of Machine Learning?
  7. 7
    Explain Supervised Learning.
  8. 8
    Explain Unsupervised Learning.
  9. 9
    Explain Semi-Supervised Learning.
  10. 10
    Explain Reinforcement Learning.
  11. 11
    Have you implemented Reinforcement Learning anywhere?
  12. 12
    Name three Unsupervised Learning algorithms.
  13. 13
    Explain your problem statement.
  14. 14
    Explain your notebook/project.
  15. 15
    How did you approach the problem?
  16. 16
    What preprocessing steps did you perform?
  17. 17
    How did you handle missing values?
  18. 18
    Did you detect outliers?
  19. 19
    How did you handle outliers?
  20. 20
    What other outlier detection methods do you know besides IQR?
  21. 21
    Explain your feature engineering.
  22. 22
    Explain your Pipeline and ColumnTransformer.
  23. 23
    Explain the charts you created.
  24. 24
    What insights would you give to the business from your analysis?
  25. 25
    What is Correlation?
  26. 26
    Which correlation method did you use?
  27. 27
    Why did you choose Pearson Correlation?
  28. 28
    Why did you select only a few features for the Correlation Matrix?
  29. 29
    Explain LightGBM.
  30. 30
    What does "Light" mean in LightGBM?
  31. 31
    Difference between XGBoost and LightGBM.
  32. 32
    Explain Random Forest.
  33. 33
    Why did you use Random Forest?
  34. 34
    What does "Forest" mean in Random Forest?
  35. 35
    Difference between Random Forest and XGBoost.
  36. 36
    What is n_estimators?
  37. 37
    How does the learning rate affect training?
  38. 38
    What happens if the learning rate changes from 0.1 to 0.01?
  39. 39
    Explain SVM.
  40. 40
    How does SVM work?
  41. 41
    Explain KNN.
  42. 42
    Explain K-Means.
  43. 43
    How does K-Means work?
  44. 44
    Difference between K-Means and K-Means++.
  45. 45
    What are the steps of the K-Means algorithm?
  46. 46
    What is PCA?
  47. 47
    How does PCA work?
  48. 48
    Is PCA supervised or unsupervised?
  49. 49
    How does PCA reduce features?
  50. 50
    How does PCA affect model performance?
  51. 51
    What is the optimal number of principal components?
  52. 52
    What are tree-based algorithms?
  53. 53
    How is the root node selected in a Decision Tree?
  54. 54
    What is Entropy?
  55. 55
    What is Information Gain?
  56. 56
    Relationship between Entropy and Information Gain.
  57. 57
    What is Gini Impurity?
  58. 58
    What is a pure node?
  59. 59
    What is an impure node?
  60. 60
    What is the height of a Decision Tree?
  61. 61
    What is the depth of a Decision Tree?
  62. 62
    Difference between a Tree and a Graph.
  63. 63
    What is Overfitting?
  64. 64
    How do you prevent Overfitting?
  65. 65
    What is Regularization?
  66. 66
    How does Regularization work?
  67. 67
    Explain L1 and L2 Regularization.
  68. 68
    What is Gradient Descent?
  69. 69
    Difference between Batch Gradient Descent and Stochastic Gradient Descent.
  70. 70
    What is the Loss Function?
  71. 71
    Loss function of Logistic Regression.
  72. 72
    Loss function of Random Forest.
  73. 73
    Loss function of Ridge Regression.
  74. 74
    Why is Mean Squared Error (MSE) used?
  75. 75
    What is R² Score?
  76. 76
    What is Precision?
  77. 77
    What is Recall?
  78. 78
    What is F1-Score?
  79. 79
    What is F1 Macro?
  80. 80
    Write the formula for F1-Score.
  81. 81
    What is a Confusion Matrix?
  82. 82
    What is TF-IDF?
  83. 83
    Write the TF-IDF formula.
  84. 84
    What are Activation Functions?
  85. 85
    Name different Activation Functions.
  86. 86
    What is ReLU?
  87. 87
    What is the input and output range of ReLU?
  88. 88
    How many hidden layers can a Neural Network have?
  89. 89
    What happens if the number of hidden layers increases?
  90. 90
    What are optimization functions in Neural Networks?
  91. 91
    What are different types of Encoders?
  92. 92
    Why is StandardScaler used?
  93. 93
    What is n_jobs?
  94. 94
    Coding: Create a dataset using random numbers.
  95. 95
    Coding: Create a random dataset with specified dimensions.
  96. 96
    Coding: Load the Digits dataset.
  97. 97
    Coding: Print the size/dimensions of the Digits dataset.
  98. 98
    Coding: Print the output classes of the Digits dataset.
  99. 99
    Coding: Apply K-Means and find the optimal number of clusters.
  100. 100
    Coding: Plot the Elbow Curve.
  101. 101
    Coding: Apply SVM on the Digits dataset.
  102. 102
    Coding: Apply a Decision Tree on the Iris dataset.
  103. 103
    Coding: Plot the Decision Tree.
  104. 104
    Coding: Load the Wine dataset.
  105. 105
    Coding: Split the dataset into training and testing sets.
  106. 106
    Coding: Train a Random Forest model.
  107. 107
    Coding: Find the model accuracy.
  108. 108
    Coding: Apply PCA before training.
  109. 109
    Coding: Compare accuracy and training time before and after PCA.
  110. 110
    Coding: Plot Accuracy vs. Number of Principal Components.
  111. 111
    Coding: Explain the PCA performance graph.
  112. 112
    Coding: Load a news article and generate the TF-IDF matrix.
  113. 113
    Coding: Stack a dataset using hstack.
  114. 114
    Coding: Use mutual_info_regression to find the top 5 features.
level1_117_MLP_Viva MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Explain your notebook/project.
  3. 3
    Why did you apply a power transformation to the target variable?
  4. 4
    Why did you choose that specific power for the transformation?
  5. 5
    Explain XGBoost.
  6. 6
    Difference between XGBoost and Extra Trees Regressor.
l1_14 MLP Level 1 Aug 7, 2026
  1. 1
    Explain your notebook/project.
  2. 2
    Explain each section of your notebook.
  3. 3
    How did you use Pipeline?
  4. 4
    What is Cross Validation (CV)?
  5. 5
    How does Cross Validation work?
  6. 6
    What resources did you use?
  7. 7
    Which is your best model?
  8. 8
    Why is it your best model?
  9. 9
    What is Hyperparameter Tuning?
  10. 10
    Which Hyperparameter Tuning technique did you use?
  11. 11
    Explain Bagging.
  12. 12
    Explain Boosting.
  13. 13
    Explain the Hyperparameter Tuning section.
  14. 14
    What is the learning rate?
  15. 15
    Where and why is the learning rate used?
  16. 16
    Coding: Load the dataset.
  17. 17
    Coding: Filter rows where ManufactureYear > 2005.
  18. 18
    Coding: For the filtered dataset, find the distribution of the UtilizationTier column.
  19. 19
    Suggestion: Add insights for each model using graphs or tables.
  20. 20
    Suggestion: Revise MLT theory and practice live coding for Level 2.
Level3_31 MLP Level 1 Aug 7, 2026
  1. 1
    Introduction.
  2. 2
    Tell me about yourself.
  3. 3
    Explain the problem statement.
  4. 4
    Explain your notebook/project.
  5. 5
    Explain why you performed each feature engineering step.
  6. 6
    Explain your EDA.
  7. 7
    Explain any graph you created.
  8. 8
    Which charts did you use and why?
  9. 9
    When should each type of chart be used?
  10. 10
    Explain your Pipeline.
  11. 11
    Explain your best model.
  12. 12
    Explain any model.
  13. 13
    How does the model work?
  14. 14
    What is Hyperparameter Tuning?
  15. 15
    Why is Hyperparameter Tuning required?
  16. 16
    Difference between GridSearchCV and RandomizedSearchCV.
  17. 17
    Explain the hyperparameters you tuned.
  18. 18
    What is the validation size in train_test_split?
  19. 19
    What is Mean, Median, and Mode?
  20. 20
    When should Mean, Median, and Mode be used for imputation?
  21. 21
    What is SimpleImputer?
  22. 22
    How do encoders work?
  23. 23
    Explain the encoding technique you used.
  24. 24
    What is Tokenization?
  25. 25
    Why do we remove stop words?
  26. 26
    Can custom stop words be used?
  27. 27
    What is a Word Cloud?
  28. 28
    Where is a Word Cloud useful?
  29. 29
    What is a Confusion Matrix?
  30. 30
    Explain F1-Score.
  31. 31
    What is the Loss Function?
  32. 32
    What is RMSE?
  33. 33
    Explain Bagging.
  34. 34
    Explain Boosting.
  35. 35
    Explain KNN.
  36. 36
    Explain K-Means.
  37. 37
    Difference between Supervised and Unsupervised Learning.
  38. 38
    What is overfitting?
  39. 39
    What is underfitting?
  40. 40
    Did you face overfitting or underfitting in your project?
  41. 41
    How did you handle it?
  42. 42
    What are outliers?
  43. 43
    How do you detect outliers?
  44. 44
    How do you handle outliers?
  45. 45
    Which visualization library did you use?
  46. 46
    Difference between Matplotlib and Seaborn.
  47. 47
    How many Kaggle submissions did you make?
  48. 48
    What changes improved your leaderboard score?
  49. 49
    How did you improve your project over different versions?
  50. 50
    Did you use AI tools during the project?
  51. 51
    How did you use AI?
  52. 52
    If you get stuck while building a model or debugging code, how would you solve it?
  53. 53
    Scenario: You are given a bank dataset to predict whether a customer will repay a credit bill. Explain your complete approach.
  54. 54
    Which model would you choose for that problem and why?
  55. 55
    Are you a student or a working professional?
  56. 56
    What are your future plans?
  57. 57
    Are you planning for higher studies?
  58. 58
    Are you participating in hackathons?
  59. 59
    Discuss your internships and projects.
level2_124 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Introduction.
  3. 3
    Tell me about yourself.
  4. 4
    How was your experience with the project/course?
  5. 5
    Show your Kaggle leaderboard score and rank.
  6. 6
    Explain the problem statement.
  7. 7
    Explain your notebook/project.
  8. 8
    How can you improve your model accuracy further?
  9. 9
    How do you know your model is not overfitting or underfitting?
  10. 10
    How did you encode categorical features?
  11. 11
    Why did you use Label Encoding instead of One-Hot Encoding?
  12. 12
    What is data preprocessing?
  13. 13
    Why is preprocessing necessary?
  14. 14
    What is Hyperparameter Tuning?
  15. 15
    Why is Hyperparameter Tuning important?
  16. 16
    Difference between GridSearchCV and RandomizedSearchCV.
  17. 17
    Explain Logistic Regression.
  18. 18
    Why is Logistic Regression called "Regression" even though it is used for classification?
  19. 19
    What is the loss function of Logistic Regression?
  20. 20
    What does the C parameter in Logistic Regression represent?
  21. 21
    Explain Decision Tree.
  22. 22
    Where are Decision Trees used in real life?
  23. 23
    What is KNN?
  24. 24
    Why is KNN called a lazy learner?
  25. 25
    Difference between KNN and K-Means.
  26. 26
    Explain the K-Means algorithm.
  27. 27
    Difference between XGBoost and LightGBM.
  28. 28
    Why did you use CatBoost?
  29. 29
    Difference between Bagging and Boosting.
  30. 30
    Which should be preferred when the model is overfitting?
  31. 31
    Explain RFE.
  32. 32
    Explain evaluation metrics such as Accuracy, Precision, Recall, F1-Score, R² Score, and Confusion Matrix.
  33. 33
    How did you conclude that your model is not overfitting?
  34. 34
    What are the limitations of a Correlation Matrix?
  35. 35
    How can you detect non-linear relationships between features?
  36. 36
    Coding: Load a dataset.
  37. 37
    Coding: Show the first few rows of the dataset.
  38. 38
    Coding: Count the number of ? values in specified columns.
  39. 39
    Coding: Replace ? with NaN.
  40. 40
    Coding: Impute or remove missing values.
  41. 41
    Coding: Convert object/string columns to numeric.
  42. 42
    Coding: Filter rows based on given conditions.
  43. 43
    Coding: Find the correlation between two columns.
  44. 44
    Coding: Plot a correlation heatmap.
  45. 45
    Coding: Normalize selected columns using MinMaxScaler or StandardScaler.
  46. 46
    Coding: Perform RFE and select the top 2 features.
  47. 47
    Coding: Perform preprocessing before applying RFE.
  48. 48
    Coding: Add two columns (e.g., V1 + V2).
  49. 49
    Coding: Perform a train-test split on the Breast Cancer dataset.
  50. 50
    Coding: Load the Breast Cancer dataset and print its shape.
  51. 51
    Coding: Load the Iris dataset and perform RFE.
  52. 52
    Coding: Explain the RFE implementation after writing the code.
level1_9 MLP Level 1 Aug 7, 2026
  1. 1
    Introduction.
  2. 2
    Explain your data preprocessing.
  3. 3
    Explain the models you used.
  4. 4
    Why did your model perform well?
  5. 5
    What is Boosting?
  6. 6
    What are the important hyperparameters of Gradient Boosting Classifier (GBC)?
  7. 7
    What is Bagging?
  8. 8
    Explain One-Hot Encoding (OHE).
  9. 9
    Explain Ordinal Encoding.
  10. 10
    Why do we use sparse_output=False in OneHotEncoder?
  11. 11
    Explain your encoding.
  12. 12
    Explain your scaling.
  13. 13
    Explain your hyperparameters.
  14. 14
    Explain your feature engineering.
  15. 15
    Coding: Separate numerical and categorical columns.
  16. 16
    Coding: Create separate preprocessing pipelines for numerical and categorical columns.
  17. 17
    Show your application/project.
  18. 18
    Coding: Modify a query using a filter.
  19. 19
    What is the difference between Authentication and Authorization?
viva_leve1_viva2 MLP Level 1 Aug 7, 2026
  1. 1
    Explain your notebook/project.
  2. 2
    Provide an in-depth explanation of your preprocessing.
  3. 3
    Provide an in-depth explanation of your feature engineering.
  4. 4
    Coding: Build the preprocessing Pipeline used in your notebook.
Level3_viva20 MLP Level 1 Aug 7, 2026
  1. 1
    Explain notebook
  2. 2
    Explain how PCA works
  3. 3
    Did you do hyper parameter tuning
Viva18 MLP Level 1 Aug 7, 2026
  1. 1
    Show your ID card.
  2. 2
    Explain your problem statement.
  3. 3
    Explain your notebook/project.
  4. 4
    How did you approach the problem?
  5. 5
    Explain your EDA.
  6. 6
    What did you show in the graphs?
  7. 7
    Explain the skewness of the target variable.
  8. 8
    Explain your preprocessing steps.
  9. 9
    Explain your feature engineering.
  10. 10
    Explain your model training.
  11. 11
    Which model did you use?
  12. 12
    How can you improve model performance?
  13. 13
    Explain your hyperparameter tuning.
  14. 14
    Show your hyperparameter tuning implementation.
  15. 15
    Explain your parameter grid.
  16. 16
    What do these hyperparameters mean? colsample_bytree learning_rate max_depth n_estimators subsample
  17. 17
    What happens if the learning rate is increased?
  18. 18
    What happens if the number of estimators is increased?
  19. 19
    What happens if the maximum depth is changed?
  20. 20
    Show your Pipeline.
  21. 21
    What is TSCV (Time Series Cross Validation)?
  22. 22
    Why did you use TSCV?
  23. 23
    What does the cv parameter mean in TSCV?
  24. 24
    Why can your dataset be considered time-series data?
  25. 25
    Why do we perform Cross Validation?
  26. 26
    Why did you use the rolling mean feature?
  27. 27
    What data/features did you not use, and why?
  28. 28
    What is the shape of the training dataset?
  29. 29
    What is the shape of the test dataset?
  30. 30
    Print the first five rows of the training dataset.
  31. 31
    Show the basic statistics of the training dataset using describe().
  32. 32
    Count the null values in the training dataset.
Most Asked Topics

No tags yet.

View all →
Top Proctors
Created for educational purposes only. Questions are based on students' personal experiences and may not reflect actual exam content.