Interactive AI & Algorithmic Systems May 25, 2025 Published project
Gomoku Master

Interactive strategy search system

This project implements Gomoku as a playable strategy game with an AI opponent. It focuses on board-state representation, search-based move selection, configurable difficulty, and practical game-loop design for an interactive decision-making system.

PythoneasyAINegamaxAlpha-beta pruningSSS*NumPyMatplotlib

Challenge

  • A playable board-game AI needs to evaluate many possible future states while staying responsive to the player.
  • Difficulty control should make the system usable for different experience levels.
  • The implementation needs a clear game loop, valid-move handling, and strategic scoring rather than random action selection.

System architecture

Board statestones and valid moves
Search engineNegamax and SSS*
Scoringline-based evaluation
Move outputdifficulty-aware play

Data and inputs

  • The main input is the current Gomoku board state and the player’s move choices.
  • The game supports multiple board sizes, including 9x9, 13x13, 15x15, and 19x19.
  • The AI uses difficulty levels to control the search depth and decision behavior.

Technical approach

  • Implement Gomoku rules for five-in-a-row wins across horizontal, vertical, and diagonal lines.
  • Use easyAI-based game logic with Negamax search, alpha-beta pruning, and transposition tables.
  • Support an additional SSS* search option for alternative game-tree exploration.
  • Score board positions according to line strength so the AI can prioritize strategic patterns.

Evaluation and results

Key indicators

4 configurable board sizes

Key indicators

5 difficulty levels

Key indicators

2 supported AI search algorithms

  • The implementation supports interactive play against an AI opponent with configurable difficulty.
  • Search-based decision logic makes the opponent evaluate board positions instead of selecting moves randomly.
  • Testing and benchmarking support continued improvement of the game engine.

Implementation and code

Implementation focus

The implementation connects data preparation, modeling, evaluation, and interpretation in a structured workflow that makes the technical decisions clear.

Source code

The code is available for exploring the implementation details and extending the experiment when needed.

Open source code

Scope and responsible use

The project demonstrates game-AI search and interactive implementation. Larger boards or stronger difficulty settings may require search-cost optimization and additional usability testing.

Future development

  • Improve the graphical interface and player experience.
  • Expand benchmarking across algorithms and difficulty levels.
  • Add richer game modes, saved games, and additional strategy diagnostics.

Technical contribution

The project demonstrates algorithmic AI design for interactive systems: representing a state space, evaluating decisions, controlling difficulty, and keeping the experience playable.