Machine Learning Concepts & Property Data

A subset of AI that develops algorithms to enable computers to learn from data without explicit programming for each task.

A computer program's performance improves on a task (T) based on experience (E) and measured by a performance measure (P).

Adjusting to changing situations and conditions.

Making choices involving multiple factors and uncertainties.

Applying learned knowledge to new situations and combining different types of learning.

Working with large amounts of data effectively.

Dealing with situations where the outcome is not clear.

Decreasing the need for human input in tasks.

Making choices based on immediate factors and circumstances.

The specific job or objective that the learning system aims to perform.

Data used to train a machine learning model.

A type of machine learning where the model is trained on labeled data.

A model that learns patterns from training data.

Properties of data used for prediction.

The correct answer for a feature in training data.

Data used to evaluate a trained model's performance.

The name or identifier of a property.

The value of the property.

A type of machine learning where the model learns by interacting with its environment and receiving rewards or penalties based on its actions.

A series of steps involved in building and deploying a machine learning solution, from problem definition to model deployment.

A type of supervised learning task that categorizes data into predefined classes or labels.

A type of supervised learning task that predicts a continuous output value based on input features.

A type of unsupervised learning task that groups data points into clusters based on their similarities.

A technique in unsupervised learning that simplifies data by reducing the number of input features while preserving important information.

Dividing customers into distinct groups based on shared characteristics, like spending habits or demographics, to tailor marketing strategies and enhance customer experience.

A collection of data points sharing similar characteristics, formed through the clustering process. It represents a distinct segment within the data.

Dividing an image into regions based on similar colors or textures, using pixel values, color histograms, and texture features.

Identifying communities or groups within a social network based on connections, shared interests, and interactions between individuals.

Grouping similar documents together based on their textual content and word frequencies, for discovering topics.

Grouping genes with similar expression patterns across samples, based on gene expression levels under different conditions.

Identifying patterns and associations in customer purchasing habits by analyzing items bought together in transactions.

An unsupervised learning technique that reduces the dimensionality of data by finding the principal components (directions of highest variance), which capture the most significant patterns.

An unsupervised learning algorithm that groups data points into a specified number of clusters (K), based on their similarity in distance to the cluster centers.

The initial steps in the machine learning pipeline focus on identifying and acquiring all data sources. It involves cleaning and organizing the data to be used for training and testing the model.

Choosing a machine learning model that best suits the problem and using the prepared data to train it. It also includes fine-tuning the model parameters for optimal performance.

The process of transforming raw data into a usable format for machine learning algorithms. It includes cleaning, structuring, and preprocessing the data.

The first stage of the data preparation process, involves identifying the data sources relevant to the task and collecting the data.

Evaluating how well the trained model performs and preparing it for real-world implementation. It involves testing the model on unseen data and deploying it for practical use.

The ethical and transparent use of AI systems, ensuring they promote societal benefits and minimize negative impacts. It focuses on reducing bias, promoting fairness, and ensuring accountability.

Responsible AI is important to minimize risks and negative impacts of AI, reduce bias, promote fairness, and increase transparency. It helps to avoid consequences for users, data subjects, and society as a whole.