Data Science and AI FBLA Test

Mean

The average of a dataset (sum divided by count).

Median

The middle value in an ordered dataset.

Mode

The most frequent value in a dataset.

Range

The difference between the highest and lowest values.

Variance

A measure of data spread from the mean.

Standard deviation

The square root of variance; typical distance from mean.

Covariance

Measures how two variables change together.

Gaussian distribution

A normal bell-shaped distribution where mean = median = mode.

Expected value

The sum of each outcome times its probability.

Continuous variable

A variable that can take any real value in a range.

Discrete variable

A variable with countable values.

Histogram

A chart that shows the distribution of data using bins.

Scatterplot

A graph showing the relationship between two numeric variables.

Boxplot

A visual of median, quartiles, and outliers.

Appropriate visual medium

Choosing the best chart to represent data clearly.

Multivariate data

Data involving more than two variables.

Multiple linear regression

Predicting a numeric value using multiple variables.

Logistic regression

Predicting categories (yes/no).

Data cleaning

Fixing issues like duplicates, missing values, or formatting errors.

Data quality issues

Problems like incomplete, duplicate, or low-quality data.

k-means

An unsupervised algorithm that creates clusters.

Decision tree

A model that predicts using rule-based splits.

Linear regression

A model that fits a line to predict a numeric value.

SQL

A language used to query and manage databases.

SELECT statement

SQL command to choose columns.

WHERE clause

SQL filter that picks rows matching conditions.

Pandas

Python library for data cleaning and manipulation.

NumPy

Python library for math and arrays.

PyTorch

A deep learning framework.

Python for data wrangling

Using Python to clean and prep datasets.

R programming language

A language used heavily in statistics.

Relational database

A database storing data in tables with relationships.

Primary key

A unique identifier for each row in a table.

Generative AI

AI that can create new content like text, images, or audio.

Capabilities of generative AI

Summaries, translations, coding, image generation.

Limitations of generative AI

Bias, hallucinations, inaccuracies.

Uses of generative AI

Healthcare, research, digital art, productivity.

Computer vision

AI that understands images and video.

Natural language processing (NLP)

AI that processes human text and speech.

Human-computer interaction

How humans interact with AI systems.

Robotics

AI that controls or automates machines.

Large language model (LLM)

A model trained on massive text datasets.

LLM capabilities

Answering questions, reasoning, summarizing, coding.

Machine learning

AI that learns patterns from data.

Training dataset

Data the model learns from.

Validation dataset

Data used to tune hyperparameters.

Test dataset

Data used to evaluate model performance.

Supervised learning

Learning from labeled data.

Unsupervised learning

Learning from unlabeled data.

Reinforcement learning

Learning through rewards and punishments.

Neural network

A model that learns complex patterns through layers.

Decision tree algorithm

A supervised model that splits data based on rules.

Deep learning

Neural networks with many layers that learn features.

Predicate logic

A symbolic way to represent facts and relationships.

Example of predicate logic

Human(Sam), Loves(Sam, Pizza).

Logic-based reasoning

Uses strict rules and true/false logic.

Probability-based reasoning

Uses uncertainty and likelihoods.

Bayesian network

A probabilistic graph of cause-effect relationships.

Node (Bayesian network)

A variable in the network.

Edge (Bayesian network)

A directional connection showing influence.

Directed acyclic graph (DAG)

A graph with no loops.

Knowledge representation

How AI stores info to reason with.

Reasoning in AI

Drawing conclusions using logic or probability.

AI dilemmas

Ethical issues like self-driving decisions or surveillance.

Algorithmic bias

Bias inherited from unfair training data.

How AI inherits bias

Through biased data, labeling, or representation.

Security risks of LLMs

Models may leak or store sensitive info.

Privacy risks of LLMs

Sensitive data may be exposed or memorized.

Hallucinations (AI)

When AI generates false information.

Misinformation from AI

Inaccurate or misleading outputs.

Surveillance concerns

AI used to track or monitor people.

Bias in generative models

Unfair or skewed outputs caused by biased data.

Ethical AI

AI systems that are fair, safe, transparent, and responsible.

Data science

Using data to answer questions or make predictions.

Structured data

Data stored in clean tables with rows and columns.

Unstructured data

Text, images, audio without organized format.

Numeric data

Quantitative values like height or cost.

Categorical data

Labels like color, brand, category.

Binary

Base-2 number system.

Hexadecimal

Base-16 number system.

Decimal

Base-10 number system.

Binary to decimal

Convert using powers of 2.

Decimal to binary

Break number into powers of 2.

Data sources

Sensors, logs, surveys, websites.

Data wrangling

Cleaning and preparing raw data.

Data transformation

Changing data format or structure.

Data science process: ask

Define the question.

Data science process: collect

Gather the needed data.

Data science process: clean

Fix errors and prep the data.

Data science process: analyze

Explore and understand data.

Data science process: model

Build predictions or insights.

Data science process: interpret

Explain what results mean.

Data science process: communicate

Share findings clearly.