A tool for extracting chapters from Gutenberg Project Italian raw text e-books. RegEx are used to match chapter headings and extract the text between them.

This repository contains the source code for The Postmodern Generator, a tool designed to generate text that mimics the style of academic postmodern criticism. It offers customizable and extensible text generation features without relying on large language models.

This is a "literary style imitation algorithm". The primary purpose is to mimic the style and tone of the original text. It creates new content based on the input text rather than directly copying existing content. It uses Markov chains for sentence generation and the ChatGPT-API for grammar cleanup.

Turn poems into living music: NLP meets melody, creative code, and literary alchemy.

Implementation of the NACL 2016 Best paper "Unsupervised Learning for Dynamic Fictional Relationships by Iyyer et al"

Contains data of the Ficiton4 corpus and for our experiment on literary sentiment evocation

Character Network of Alfred de Musset's play Lorenzaccio

Gibson encoded cognitive mismatch in 1984, before the vocabulary existed to name it — a forensic, experiential reading of Neuromancer.

Pipeline multiagente local para análisis y evaluación editorial de manuscritos con LLMs. Corre 100% en local con LM Studio — el manuscrito nunca sale de tu máquina.

Stylometric Measurements on Gutenberg Corpora

Comparing chapter-level and sentence-level sentiment analysis

Local RAG system for analyzing James Joyce's Finnegans Wake (Chapter I.8)

Trope Miner — a local, privacy-first pipeline that mines narrative tropes from fiction using embeddings + LLMs, with review UI, span verification, semantic seeding, and calibration.