Inspired by the original Zork games. A console-based text adventure game with a full featured Natural Language Processing (NLP) implementation and use.

• .NET / C#
• Stanford CoreNLP for .NET https://sergey-tihon.github.io/Stanford.NLP.NET/StanfordCoreNLP.html

• Code base purposely over-engineered to best demonstrated my skills and knowledge in .NET and C# code design
• Adhered to SOLID design pricinples
• All movement and look intentions are recognised by the fully featured parser and actioned upon
• All other possible intentions that the player may want to do are recognised by the NLP system, but the game does not act upon them (to come in a later update)
• Adverbs and adjectives are all fully recognised and utilised - e.g. "very quickly move ..." results in a different action to "slowly move ..."
• Full Named Entity Recognition (NER) support, so points of interests like real world locations are recognised by the NLP system

The player can freely move around the environment using intuitive English phrases as commands. E.g. "go to the north", "walk to the hallway", or simply "north"

Similarly, the player can look around their environment or at points of interests or items. E.g. "look at room", "inspect the rock"

1. Clone, or download and extract this repository
2. Download both Stanford CoreNLP https://stanfordnlp.github.io/CoreNLP/download.html, as well as the English "model jar" file, and extract them both somewhere locally (use a file archiver like 7-Zip to extract the JAR file)
3. Open TextAdventure.Command/CommandBootstrap.cs and change line 70 to point to the /models subdirectory and change line 74 to point to /englishPCFG.ser.gz (found in /edu/stanford/nlp/models/lexparser subdirectory)
4. Rebuild all and run

@chullman https://github.com/chullman/