WorksheetsExploring Machine Learning Concepts
Total questions: 15
Worksheet time: 5mins
What is the primary goal of supervised learning?
To learn a mapping from input features to output labels based on labeled training data.
To predict future outcomes without any training data.
To optimize the performance of unsupervised learning algorithms.
To classify data into predefined categories without labels.
How does unsupervised learning differ from supervised learning?
Unsupervised learning requires labeled data, while supervised learning does not.
Both unsupervised and supervised learning use the same type of data.
Unsupervised learning is only applicable to classification tasks.
Unsupervised learning does not use labeled data, while supervised learning does.
What is a neural network and how does it function?
A neural network is a computational model that processes information through interconnected layers of nodes, learning by adjusting weights based on prediction errors.
A neural network is a type of biological brain that functions without any training.
A neural network is a hardware device that processes data without any learning capabilities.
A neural network is a simple algorithm that only uses linear equations to make predictions.
Explain the concept of a decision tree in machine learning.
A decision tree is a type of neural network that processes data in layers.
A decision tree is a model that uses a tree-like structure to make decisions based on input features, splitting data into subsets to predict outcomes.
A decision tree is a statistical method that calculates probabilities without any data splitting.
A decision tree is a linear model that predicts outcomes based on a single feature.
What are support vector machines used for?
Support vector machines are used for image editing.
Support vector machines are used for classification and regression tasks.
Support vector machines are primarily for data storage.
Support vector machines are used for web development.
Describe the process of model evaluation in machine learning.
The process of model evaluation in machine learning includes splitting the dataset, training the model, making predictions, and comparing results using evaluation metrics.
Visualizing the training data
Deploying the model to production
Collecting data from various sources
What does ANN stand for and what is its significance?
Automated Neural Network
Artificial Network Node
Artificial Neural Network
Analog Neural Network
How do convolutional neural networks (CNN) differ from traditional neural networks?
CNNs are only used for image processing tasks.
Traditional neural networks are faster than CNNs for all applications.
CNNs do not require any training data unlike traditional neural networks.
CNNs use convolutional layers for feature extraction, while traditional neural networks use fully connected layers.
What is the KNN algorithm and how does it work?
KNN is a non-parametric algorithm that classifies data points based on the majority class of their k nearest neighbors.
KNN uses a decision tree to classify data points based on their features.
KNN is a linear regression algorithm that predicts continuous values.
KNN is a clustering algorithm that groups data points into distinct clusters.
What are the advantages of using decision trees?
High computational cost
Limited to binary outcomes
Requires extensive data cleaning
Advantages of using decision trees include interpretability, versatility with data types, minimal preprocessing, ability to model complex relationships, and applicability to both classification and regression.
In what scenarios would you prefer using support vector machines?
When interpretability of the model is the highest priority.
When data is high-dimensional, classes are well-separated, and robustness against overfitting is needed.
When classes are overlapping and not well-defined.
When data is low-dimensional and noisy.
How can you assess the performance of a machine learning model?
Ignore the model's predictions and focus on its architecture.
Use metrics like accuracy, precision, recall, F1 score, and AUC-ROC.
Assess performance based on the model's training time.
Use only the training data for evaluation.
What role does feature scaling play in KNN?
Feature scaling reduces the number of features in KNN.
Feature scaling ensures that all features contribute equally to distance calculations in KNN.
Feature scaling eliminates the need for distance calculations in KNN.
Feature scaling is only necessary for categorical data in KNN.
What is overfitting and how can it be prevented?
Overfitting is when a model performs well on new data but poorly on training data.
Overfitting can be prevented by using more complex models and ignoring validation data.
Overfitting is when a model learns the training data too well, leading to poor performance on new data. It can be prevented by using techniques like cross-validation, regularization, and simpler models.
Overfitting occurs when a model is too simple and cannot learn the training data.
Explain the difference between classification and regression in supervised learning.
Classification predicts numerical values, while regression predicts categories.
Regression outputs discrete labels, while classification outputs continuous values.
Classification is used for time series data, while regression is for image data.
Classification deals with categorical outputs, while regression deals with continuous outputs.
