Home Courses Facebook Friend Recommendation using Graph Mining Questions & Answers

Questions & Answers

Instructor: Applied AI Course Duration: 30 mins

Revision Questions:

  1. What is Accuracy ?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/accuracy-1/)

  2. Explain about Confusion matrix, TPR, FPR, FNR, TNR?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/confusion-matrix-tpr-fpr-fnr-tnr-1/)

  3. What do you understand  about Precision & recall, F1-score? How would you use it?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/precision-and-recall-1/)

  4. What is the ROC Curve and what is AUC (a.k.a. AUROC)?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/receiver-operating-characteristic-curve-roc-curve-and-auc-1/)

  5. What is Log-loss and how it helps to improve performance?.(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/log-loss-1/)

  6. Explain about R-Squared/ Coefficient of determination.(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/r-squared-1/)

  7. Explain about Median absolute deviation (MAD) ?Importance of MAD?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/median-absolute-deviation-mad-1/)

  8. Define Distribution of errors?(https://www.appliedaicourse.com/course/applied-ai-course-online/lessons/distribution-of-errors/)

Self Learning:

  1. Which is more important to you– model accuracy, or model performance?

  2. Can you cite some examples where a false positive is important than a false negative?

  3. Can you cite some examples where a false negative important than a false positive?

  4. Can you cite some examples where both false positive and false negatives are equally important?

  5. What is the most frequent metric to assess model accuracy for classification problems?

  6. Why is Area Under ROC Curve (AUROC) better than raw accuracy as an out-of- sample evaluation metric?

**If you face any new Interview questions please put in comments ,we will work it out**

Prev
Next
Distribution of errors
Conditional probability

0 Comment(s)

Login to comment
1.1
Dataset overview: Amazon Fine Food reviews(EDA)
23 min
1.2
Data Cleaning: Deduplication
15 min
1.3
Why convert text to a vector?
14 min
1.4
Bag of Words (BoW)
18 min
1.5
Text Preprocessing: Stemming, Stop-word removal, Tokenization, Lemmatization.
15 min
1.6
uni-gram, bi-gram, n-grams.
9 min
1.7
tf-idf (term frequency- inverse document frequency)
22 min
1.8
Why use log in IDF?
14 min
1.9
Word2Vec.
16 min
1.10
Avg-Word2Vec, tf-idf weighted Word2Vec
9 min
1.11
Bag of Words( Code Sample)
19 min
1.12
Text Preprocessing( Code Sample)
11 min
1.13
Bi-Grams and n-grams (Code Sample)
5 min
1.14
TF-IDF (Code Sample)
6 min
1.15
Word2Vec (Code Sample)
12 min
1.16
Avg-Word2Vec and TFIDF-Word2Vec (Code Sample)
2 min
1.17
Exercise: t-SNE visualization of Amazon reviews with polarity based color-coding
6 min
2.1
How “Classification” works?
10 min
2.2
Data matrix notation
7 min
2.3
Classification vs Regression (examples)
6 min
2.4
K-Nearest Neighbours Geometric intuition with a toy example
11 min
2.5
Failure cases of KNN
7 min
2.6
Distance measures: Euclidean(L2) , Manhattan(L1), Minkowski, Hamming
20 min
2.7
Cosine Distance & Cosine Similarity
19 min
2.8
How to measure the effectiveness of k-NN?
16 min
2.9
Test/Evaluation time and space complexity
12 min
2.10
KNN Limitations
9 min
2.11
Decision surface for K-NN as K changes
23 min
2.12
Overfitting and Underfitting
12 min
2.13
Need for Cross validation
22 min
2.14
K-fold cross validation
17 min
2.15
Visualizing train, validation and test datasets
13 min
2.16
How to determine overfitting and underfitting?
19 min
2.17
Time based splitting
19 min
2.18
k-NN for regression
5 min
2.19
Weighted k-NN
8 min
2.20
Voronoi diagram
4 min
2.21
Binary search tree
16 min
2.22
How to build a kd-tree
17 min
2.23
Find nearest neighbours using kd-tree
13 min
2.24
Limitations of Kd tree
9 min
2.25
Extensions
3 min
2.26
Hashing vs LSH
10 min
2.27
LSH for cosine similarity
40 min
2.28
LSH for euclidean distance
13 min
2.29
Probabilistic class label
8 min
2.30
Code Sample:Decision boundary .
23 min
2.31
Code Sample:Cross Validation
13 min
2.32
Exercise: Apply k-NN on Amazon reviews dataset
5 min
3.1
Questions & Answers

30 min
4.1
Introduction
5 min
4.2
Imbalanced vs balanced dataset
23 min
4.3
Multi-class classification
12 min
4.4
k-NN, given a distance or similarity matrix
9 min
4.5
Train and test set differences
22 min
4.6
Impact of outliers
7 min
4.7
Local outlier Factor (Simple solution :Mean distance to Knn)
13 min
4.8
k distance
4 min
4.9
Reachability-Distance(A,B)
8 min
4.10
Local reachability-density(A)
9 min
4.11
Local outlier Factor(A)
21 min
4.12
Impact of Scale & Column standardization
12 min
4.13
Interpretability
12 min
4.14
Feature Importance and Forward Feature selection
22 min
4.15
Handling categorical and numerical features
24 min
4.16
Handling missing values by imputation
21 min
4.17
curse of dimensionality
27 min
4.18
Bias-Variance tradeoff
24 min
4.19
Intuitive understanding of bias-variance.
6 min
4.20
best and wrost case of algorithm
6 min
5.1
Accuracy
15 min
5.2
Confusion matrix, TPR, FPR, FNR, TNR
25 min
5.3
Precision and recall, F1-score
10 min
5.4
Receiver Operating Characteristic Curve (ROC) curve and AUC
19 min
5.5
Log-loss
12 min
5.6
R-Squared/Coefficient of determination
14 min
5.7
Median absolute deviation (MAD)
5 min
5.8
Distribution of errors
7 min
7.1
Conditional probability
13 min
7.2
Independent vs Mutually exclusive events
6 min
7.3
Bayes Theorem with examples
18 min
7.4
Exercise problems on Bayes Theorem
30 min
7.5
Naive Bayes algorithm
26 min
7.6
Toy example: Train and test stages
26 min
7.7
Naive Bayes on Text data
16 min
7.8
Laplace/Additive Smoothing
24 min
7.9
Log-probabilities for numerical stability
11 min
7.10
Bias and Variance tradeoff
14 min
7.11
Feature importance and interpretability
10 min
7.12
Imbalanced data
14 min
7.13
Outliers
6 min
7.14
Missing values
3 min
7.15
Handling Numerical features (Gaussian NB)
13 min
7.16
Multiclass classification
2 min
7.17
Similarity or Distance matrix
3 min
7.18
Large dimensionality
2 min
7.19
Best and worst cases
8 min
7.20
Code example
7 min
7.21
Exercise: Apply Naive Bayes to Amazon reviews
6 min
8.1
Geometric intuition of Logistic Regression
31 min
8.2
Sigmoid function: Squashing
37 min
8.3
Mathematical formulation of Objective function
24 min
8.4
Weight vector
11 min
8.5
L2 Regularization: Overfitting and Underfitting
26 min
8.6
L1 regularization and sparsity
11 min
8.7
Probabilistic Interpretation: Gaussian Naive Bayes
19 min
8.8
Loss minimization interpretation
19 min
8.9
hyperparameters and random search
16 min
8.10
Feature importance and model interpretability
14 min
8.11
Collinearity of features
14 min
8.12
Test/Run time space and time complexity
10 min
8.13
Real world cases
11 min
8.14
Non-linearly separable data & feature engineering
28 min
8.15
Column Standardization
5 min
8.16
Code sample: Logistic regression, GridSearchCV, RandomSearchCV
23 min
8.17
Exercise: Apply Logistic regression to Amazon reviews dataset.
6 min
8.18
Extensions to Generalized linear models
9 min
9.1
Geometric intuition
13 min
9.2
Mathematical formulation
14 min
9.3
Real world Cases
8 min
9.4
Code sample for Linear Regression
13 min
10.1
Differentiation
29 min
10.2
Online differentiation tools
8 min
10.3
Maxima and Minima
12 min
10.4
Vector calculus: Grad
10 min
10.5
Gradient descent: geometric intuition
19 min
10.6
Learning rate
8 min
10.7
Gradient descent for linear regression
8 min
10.8
SGD algorithm
9 min
10.9
Constrained Optimization & PCA
14 min
10.10
Logistic regression formulation revisited
6 min
10.11
Why L1 regularization creates sparsity?
17 min
10.12
Exercise: Implement SGD for linear regression
6 min
10.13
Interview Questions on Logistic Regression and Linear Regression

30 min
11.1
Questions & Answers

30 min