github.com/AndyFerns/Automated-Reasoning-Project ↗

About AndyFerns/Automated-Reasoning-Project

This project implements an automated reasoning engine capable of performing logical inference on first-order logic statements using natural language processing. The system bridges the gap between human-readable text and formal logic by leveraging NLP techniques to convert natural language into predicate logic forms similar to Prolog statements. It extracts logical constructs from text, identifies atomic sentences, and performs reasoning based on user-defined rules.

The engine combines several NLP techniques including tokenization and part-of-speech tagging using NLTK, proper noun classification to distinguish between different noun types, and atomic sentence detection to construct logical predicates. The architecture emphasizes modularity, making it accessible for understanding, debugging, and extending the codebase. Users can prepare input text or logical statements and run the system to obtain parsed predicates and inferred conclusions.

The project targets developers and researchers interested in AI-driven reasoning systems, logical inference, and natural language understanding. It demonstrates how classical logic programming concepts can be integrated with modern NLP libraries like spaCy to create a flexible framework for automated reasoning applications, though the current implementation appears to be in development with example code snippets referenced but not fully detailed in the documentation.