Pride & Prejudice & C#
summary
C# Markov chains read Pride and Prejudice from Project Gutenberg, split into sentences, replaced punctuation with spaced tokens, and treat 2-word state pairs as states stored in a dictionary with counts converted to repeated-choice lists. The generator runs until a set number of paragraph breaks (often 500) and produces megabytes of Austen‑style text that mimics punctuation and word flavor but lacks plot. Tried 3‑ and 4‑word chains (they mostly regurgitate source), and mixed in Trump tweets to show domain quirks; works for demo and testing, not long‑term coherence.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 0:43Modeling cereal choices
- 6:02What are Markov chains
- 7:28Mathematical representation
- 12:05Markov chains in Austen's writing
- 15:38Text generation process
- 21:38Generating tweets
- 23:44Real-world uses
- 27:17New Austen novels
- 28:04Q&A
related talks
40mText similiarity with the next generation of word embeddings in Gensim
24mPyambic Pentameter: generating rhyming and metered poems
30mRuby-us Hagrid: Writing Harry Potter with Ruby
11mMutating text toward better nonsense?!
33mSyntax Isn't Everything: NLP for Rubyists
36mProbabilistic scripts for automating common-sense tasks
26mAnalytic Pattern Matching: From DNA to Twitter
04mLightning Talks – Przemyslaw Krowinski