IS 4490 - Unstructured Data, Preprocessing, and Machine Learning Type

Which of the following is NOT an example of unstructured data?

Relational database tables

What is feature engineering?

The process of using domain knowledge to extract features from raw data

Which term is often used interchangeably with "target" in supervised learning?

Dependent variable (DV)

Which type of classification problem involves predicting exactly two classes?

Binary Classification

In multilabel classification, what is unique about the labeling process?

Each instance may be assigned multiple labels

Which of the following is a common application of regression?

Stock price prediction

Which preprocessing technique is commonly used for unstructured text data?

Natural language processing

What is the purpose of data preprocessing?

To clean and transform raw data into a suitable format for machine learning models.

According to the text, which type of machine learning uses a labeled dataset to train an algorithm to predict a target value?

Supervised learning

What is the primary difference between classification and regression?

Classification predicts a discrete label, while regression predicts a continuous numerical value.

What is a key advantage of deep learning models regarding feature engineering?

They can automatically learn complex features from raw data.

In K-means clustering, the algorithm starts with predefined cluster centroids that never change.

False

Association rules only work with purchase data and cannot be applied to other domains.

False

Reinforcement learning uses a labeled dataset to train an agent.

False

Shallow machine learning models typically use multiple layers of neurons.

False

Convolutional Neural Networks (CNNs) use convolutional layers to learn spatial hierarchies of features.

True