Data Mining Questions et reponses d'entretien
Question 26. How does the naive Bayes classifier work in data mining?
Naive Bayes is a probabilistic classification algorithm based on Bayes' theorem. It assumes independence between features and calculates the probability of a class given the input features.
Example:
Classifying emails as spam or non-spam based on the occurrence of words in the email content.
Question 27. What is the role of a confusion matrix in evaluating classification models?
A confusion matrix summarizes the performance of a classification model by showing the number of true positive, true negative, false positive, and false negative predictions.
Example:
Evaluating a binary classifier's performance in predicting disease outcomes.
Question 28. What is the concept of imbalanced datasets, and how does it impact machine learning models?
Imbalanced datasets have unequal distribution of classes, leading to biased models. It can result in poor performance on the minority class and overfitting on the majority class.
Example:
A fraud detection model trained on a dataset where only 1% of transactions are fraudulent.
Question 29. Explain the difference between feature extraction and feature engineering.
Feature extraction involves transforming raw data into a new representation, while feature engineering involves creating new features or modifying existing ones to improve model performance.
Example:
Feature extraction: Using PCA to reduce dimensionality. Feature engineering: Creating a new feature by combining existing ones.
Question 30. What is the purpose of cross-validation in machine learning, and how does it work?
Cross-validation is a technique used to assess a model's performance by splitting the dataset into multiple subsets. It helps provide a more accurate estimate of how the model will generalize to unseen data by training and evaluating the model on different subsets in multiple iterations.
Example:
Performing 5-fold cross-validation involves dividing the dataset into five subsets. The model is trained on four subsets and tested on the remaining one, repeating the process five times with a different test subset each time.
Les plus utiles selon les utilisateurs :