This repository contains the source code for the SCFirstOrderLogic NuGet package - along with its tests, performance benchmarks, and documentation website. It also contains some example domains and alternative algorithm implementations, which are used in the tests and performance benchmarks.
For documentation of the package itself, see https://sdcondon.net/SCFirstOrderLogic/.
I have not written up any documentation of the source (e.g. repo summary, design discussion, build guidance) - and likely won't unless someone else expresses an interest in contributing. Once cloned, it should "just work" as far as compilation is concerned. The only thing perhaps worthy of note is that it uses Antlr4BuildTasks to invoke ANTLR to generate the FoL-parsing code. ANTLR is a Java tool - something Antlr4BuildTasks handles by downloading the JRE to your system for you - see its docs for details.
I'm not expecting anyone to want to get involved given the low download figures on NuGet, but please feel free to do so. I do keep an eye on the issues tab, and will add a CONTRIBUTING.md if anyone drops me a message expressing interest. Do bear in mind that I have a very particular scope in mind for the library, though - see the roadmap for details.
You might also be interested in:
- Some related AI projects:
- SCClassicalPlanning: Basic classical planning implementations. Based on chapter 10 of "Artificial Intelligence: A Modern Approach" - though perhaps a little more loosely than this repo is based on chapters 8 and 9. Specifically, introduces a couple more PDDL concepts that the book doesn't bother with (but should IMO). Depends on this library (and SCGraphTheory.Search).
- SCPropositionalLogic: Basic propositional logic implementations. Based on chapter 7 of "Artificial Intelligence: A Modern Approach". Mostly just a precursor to this repo - I haven't bothered publishing this one to NuGet - but might be of use if you find this repo a bit too much.
- Some unrelated AI projects:
- Infer.NET: Bayesian inference engine. Bayesian logic of course being strictly more powerful than first-order logic.
- AIMA Code - C#: I mention this only because I feel like I should. This is the "official" C# repository for "Artificial Intelligence: A Modern Approach" - and it is.. err.. not great.