Machine Learning Interview Questions

Machine Learning Interview Question 

  1. Explain Machine learning

Ans. It is a subdomain of Artificial Intelligence.

Machine learning is a branch of computer science which deals with computer programming that automatically improve their performance through experience.

  1. Enlist the categories of machine learning algorithm?

Ans.

  • Supervised Learning
  • Unsupervised Learning
  • Semi-supervised Learning
  • Reinforcement Learning
  1. What is classifier in machine learning?

Ans. A classifier is a system that inputs a vector of discrete or continuous feature values and outputs a single discrete value.

  1. What is supervised learning?

Ans. Supervised learning is a process of machine learning in which outputs are fed back into a computer for the system to learn from for more accurate results the next time.

Name itself indicates that there is a presence of a supervisor as a teacher in supervised learning.

  1. Algorithms fall into the categories of supervised learning.

Ans.

  • Classification
  • Regression

Classification: In this type the output should be in discrete terms that is it should be either yes or no. For example, spam email classifier, Tumor detection,

Regression: In this type the output should not be in discrete terms. For example, weather forecasting, stock exchange and share market prediction.

  1. What are the various techniques for regression?

Ans. 

  • Linear regression
  • Logistic regression
  • Polynomial regression
  • Stepwise regression
  • Stepwise regression
  • Ridge regression
  • Lasso regression
  • ElasticNet regression
  1. What are the examples of supervised learning algorithms?

Ans. 

  • Logistic Regression
  • Neural networks
  • Support Vector Machines (SVMs)
  • Naive Bayes classifiers.
  1. Explain Unsupervised learning

Ans. Unsupervised learning is the training of machine using information that is neither classified nor labeled and allowing the algorithm to act on that information without guidance.

  1. What are the examples of unsupervised learning algorithms?

Ans. K-means, Random Forests, Hierarchical clustering

  1. What Is Semi-Supervised Learning?

Ans. A training dataset with both labeled and unlabeled data is used in Semi-Supervised Learning.

  1. What Is Reinforcement Learning?

Ans. In Reinforcement Learning, learning data gives feedback to the system to adjusts into dynamic conditions in order to achieve a certain objective.

  1. What are the various types of problem handled in unsupervised learning?

Ans.

  • Clustering
  • Association
  • Anomaly detection
  • Autoencoders
  1. What is the difference between KNN and K-means algorithm?

Ans. K-Nearest Neighbor is a supervised classification algorithm, whereas k-means is an unsupervised clustering algorithm.

  1. What is ensemble learning?

Ans. To solve a computational program, multiple models such as classifiers or experts are strategically generated and combined. This process is known as ensemble learning.

  1. Which is called as a High Bias in machine learning?

Ans. Under-fitting

  1. Which is called as High variance in machine learning?

Ans. Over-fitting

  1. What are the various techniques to overcome Over-fitting in machine learning?

Ans. They are three types of regularization technique,

  1. a) L1 regularization (Lasso regularization)
  2. b) L2 regularization (Ridege regularization)
  3. c) Elastic net

18. What is PCA?

Ans. PCA (Principal Components Analysis), is the important feature extraction techniques used for dimensionality reduction.

  1. What is SVM?

Ans. Support vector machines (SVM) is supervised learning algorithm used for classification and regression analysis.

  1. Explain the terms TP, FP, TN and FN.

Ans.

  • True Positive (TP): Observation which is positive and is predicted to be positive.
  • False Positive (FP): Observation which is negative but is predicted positive.
  • True Negative (TN): Observation which is negative and is predicted to be negative.
  • False Negative (FN): Observation which is positive but is predicted negative.
  1. What is formula for classification accuracy?

Ans.  Accuracy = TP+TN/TP+TN+FP+FN

  1. what is Recall?

Ans. It is known as a true positive rate.

Recall is defined as the ratio of the total number of correctly classified positive samples divide to the total number of positive samples.

High the Recall value better the class recognized correctly.

  1. Write down the formula for RECALL

Ans.  Recall = TP/TP+FN

  1. What is Precision?

Ans. It is also known as a positive predicted value.