Unit 4: Statistical Data (Theory & Practical) Vocabulary

Comprehensive vocabulary flashcards covering Statistical Data, Data Science applications, High-Code/Low-Code/No-Code AI, Statistics concepts, Linear Regression, and Orange Data Mining (ODM) stages and widgets.

Data Science

A domain of AI and study of data that extracts information using modern tools, techniques, and complex machine learning algorithms from different sources or formats to derive meaning and sense out of data systems.

Click-Through Rate (CTR)

A metric in digital advertising measuring user engagement, which is higher for targeted ads based on user past behavior compared to traditional advertisements.

High-Code

Traditional software development where programmers manually write code using programming languages such as Java, Python, or C#; also known as custom-code.

Low-Code AI

An AI development approach where users with some programming skills build applications with minimum coding using visual drag-and-drop components.

No-Code AI

Tools and platforms allowing users without technical backgrounds to build AI applications using drag-and-drop interfaces without writing code.

Automation Bias

The tendency for humans to favor suggestions from automated decision-making systems and ignore contradictory information made without automation, even if it is correct.

Azure Machine Learning

A cloud-based no-code tool provided by Microsoft, released in July 2014.

Google Cloud AutoML

A cloud-based no-code service provided by Google, released in January 2018.

Orange Data Mining (ODM)

An open-source data visualization, machine learning, and data mining toolkit featuring a visual programming interface and Python scripting, developed by the University of Ljubljana and released in October 1996.

Lobe AI

A machine learning platform released in 2015 that enables users to create custom machine learning models using a visual interface.

Teachable Machine

A web-based tool released in November 2017 that makes creating machine learning models fast, easy, and accessible to everyone.

Population

The complete collection of all raw data available for a specific test or experiment.

Statistical Sampling

The practice of selecting a sample—a smaller, representative subset of the population—to save time, money, and resources when analyzing the full population is unfeasible.

Probability

The likelihood or chance of an event (the outcome of an experiment) occurring, where events are classified as independent or dependent.

Mean

The most common measure of a typical value, calculated by adding all numbers in a dataset and dividing by the count of those numbers.

Median

The middle value when data is arranged in ascending order, calculated as the center number for an odd number of values or the average of the two middle numbers for an even number of values.

Mode

The number that appears most frequently in a dataset.

Variance

A measure of how much data values differ (scatter) from the mean, categorized into high variance (widely spread) and low variance (close to the mean).

Standard Deviation

The square root of the variance, measuring data spread in the original units of the data.

Outlier

An exceptional data point that is significantly higher or lower than the rest of the values in a dataset.

Normal Distribution

A perfectly symmetrical distribution (bell curve) where the mean, median, and mode are all equal and located at the center.

Left-Skewed Distribution

A data distribution with a long tail on the left side (lower values), where the mean is usually less than the median, which is less than the mode.

Right-Skewed Distribution

A data distribution with a long tail on the right side (higher values), where most data is clustered on the left and the mean is usually greater than the median, which is greater than the mode.

Linear Regression

A statistical method used to understand the relationship between an independent variable ($X$) and a dependent variable ($Y$).

Problem Scoping

The first stage of the AI Project Cycle in ODM, involving the identification of objectives, feasibility, and data sources to align with business goals.

Data Acquisition

The second stage of the AI Project Cycle in ODM, which involves gathering and uploading the dataset.

Data Exploration

The third stage of the AI Project Cycle in ODM, focusing on understanding and cleaning data through steps like handling missing data, selecting relevant columns, and data sampling.

Modelling

The fourth stage of the AI Project Cycle in ODM, where an appropriate algorithm is selected and trained.

Evaluation

The fifth stage of the AI Project Cycle in ODM, assessing model reliability on unseen data using performance metrics and predictions.

Deployment

The sixth stage of the AI Project Cycle in ODM, implementing the trained model into a real-world environment with continuous monitoring and retraining.

Data Widgets

Modular tools in Orange Data Mining used for loading and interacting with data, including File, CSV File Import, and Data Table widgets.

Transform Widgets

Modular tools in Orange Data Mining used for preprocessing and feature engineering, such as Impute and Select Columns.

Visualize Widgets

Modular tools in Orange Data Mining used for identifying insights through charts, such as Box Plot, Scatter Plot, and Tree Viewer.

Model Widgets

Modular tools in Orange Data Mining used to train algorithms, such as Random Forest, Linear Regression, and Logistic Regression.

Evaluate Widgets

Modular tools in Orange Data Mining used to assess model effectiveness, including Test and Score, Confusion Matrix, and ROC Analysis.

Unsupervised Widgets

Modular tools in Orange Data Mining used for data without labeled outcomes, featuring tools like K-Means for clustering and t-SNE for dimensionality reduction.