A Local Sequence Alignment Approach to Recognizing Fixed Poetic Forms Across Languages

This repo contains replication data and code for "A Local Sequence Alignment Approach to Recognizing Fixed Poetic Forms Across Languages". 01_clustering.py builds local sequence alingment based distance matrix of all poems, performs UMAP dimensionality reduction and HDBSCAN clustering distance matrix is stored in data/distance_matrix.pkl, all input data are stored along their corresponding cluster labels in data/clustered.json ! distance matrix is not part of the repo and needs to be built first ! 02_visualize.ipynb provides 2D representations of clustering (section 2 of the article) 03_eval.ipynb performs clustering evaluation (section 3) 04_showcase1_donors.ipynb showcase 1, section 4.1 04_showcase2_topics.ipynb showcase 2, section 4.2 04_showcase3_locations.ipynb showcase 2.section 4.3