Dark Mode
Tic-Tac-Toe Binary Classification Database
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Encoding the complete set of possible board configurations at the end of Noughts and Crosses (Tic-Tac-Toe) games, this database serves as a resource for machine learning and logic experimentation. The data assumes 'x' plays first and targets the concept of a "win for x" (true when 'x' has achieved one of 8 possible ways to create a "three-in-a-row"). It captures the state of the board across all valid endgame scenarios, providing a straightforward dataset for testing binary classification algorithms and understanding decision trees in game theory contexts.
Columns
- top-left-square: Represents the status of the top-left position on the board (values: 'x', 'o', 'b' for blank).
- top-middle-square: Represents the status of the top-middle position on the board (values: 'x', 'o', 'b').
- top-right-square: Represents the status of the top-right position on the board (values: 'x', 'o', 'b').
- middle-left-square: Represents the status of the middle-left position on the board (values: 'x', 'o', 'b').
- middle-middle-square: Represents the status of the middle-centre position on the board (values: 'x', 'o', 'b').
- middle-right-square: Represents the status of the middle-right position on the board (values: 'x', 'o', 'b').
- bottom-left-square: Represents the status of the bottom-left position on the board (values: 'x', 'o', 'b').
- bottom-middle-square: Represents the status of the bottom-middle position on the board (values: 'x', 'o', 'b').
- bottom-right-square: Represents the status of the bottom-right position on the board (values: 'x', 'o', 'b').
- Class: The target variable indicating the game outcome for player 'x' (values: 'positive' for a win, 'negative' for no win).
Distribution
- Format: CSV (tic_tac_toc.csv)
- Size: 55.74 kB
- Structure: 10 columns
- Records: 958 valid rows (instances)
Usage
- Binary Classification: Ideal for training models to predict game outcomes based on board states.
- Decision Tree Learning: Useful for teaching and visualising simple decision tree algorithms and rule-based logic.
- Game Theory Analysis: Supports the study of finite game states and win conditions in solved games.
- Algorithm Benchmarking: Provides a clean, categorical dataset for benchmarking simple predictive algorithms.
Coverage
- Scope: The dataset encompasses the complete set of possible board configurations at the end of a Tic-Tac-Toe game.
- Demographic/Geographic: Not applicable (Abstract Logic/Game).
- Completeness: Covers 958 valid endgame states with no missing or mismatched values.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
- Data Science Students: For practising categorical data handling and classification techniques.
- Machine Learning Educators: As a clear, understandable example for teaching supervised learning concepts.
- Game Developers: For analysing state-space complexity in simple board games.
Dataset Name Suggestions
- Noughts and Crosses Endgame States
- Tic-Tac-Toe Binary Classification Database
- Complete Tic-Tac-Toe Board Configurations
- Player X Win Prediction Dataset
Attributes
Original Data Source: Tic-Tac-Toe Binary Classification Database
QUALITYLoading...