Answer / Narendra Pratap Singh

Ensemble methods combine multiple weak learners to form a strong learner. The general idea behind ensemble methods is to reduce the variance and overfitting by averaging or voting predictions from multiple models. Bagging (Bootstrapped Aggregating) creates multiple subsets of training data with replacement and trains a model on each subset, while boosting sequentially trains models that correct the errors of the previous one.

What cross-validation technique would you use on a time series dataset?

1 Answers  

---

What are convolutional networks? Where can we use them?

1 Answers  

---

Which one would you prefer to choose – model accuracy or model performance?

1 Answers  

---

Why is “naive” bayes naive?

1 Answers  

---

Why is logistic regression better than naive bayes?

1 Answers  

---

What is the classification threshold in machine learning?

1 Answers  

---

Explain what is the difference between inductive machine learning and deductive machine learning?

1 Answers  

---

What do you understand by algorithm independent machine learning?

1 Answers  

---

What are the smaller dataset techniques?

1 Answers  

---

Why do we need to convert categorical variables into factor? Which functions are used to perform the conversion?

1 Answers  

---

What is regularization in machine learning?

1 Answers  

---

List down various approaches to machine learning?

1 Answers  

---