Clustering Tasks and Decision Trees with Elegiac Poets

<p>The dataset contains files generated during a Natural Language Processing (NLP) and automatic text analysis task. Attached is a <strong>Jupyter notebook</strong> with the complete code, along with several <strong>Excel files (.xlsx)</strong> containing organized information. Additionally, there are three folders that include files generated during the Silhouette calculation, K-means clustering, and feature extraction using decision trees.</p> <p>The three folders are:<br>1. <strong>Silhouette Calculation:</strong> Contains PNG images of Silhouette plots for various analysis configurations.<br>2.<strong> K-means Clustering: </strong>Contains pickle (.pkl) files with features and labels for each combination of excluded author, n-gram type, n-gram range, and matrix type.<br>3. <strong>Feature Extraction: </strong>Contains CSV files with lists of documents by cluster and the most important features along with information gain and information gain ratio metrics.</p> <p>Other file formats included in the dataset are:<br>- CSV files containing Silhouette scores, optimal clustering results, cluster assignments, and optimal cluster assignments.<br>- PNG images of scatter plots colored by author and by cluster.<br>- Pickle files containing the top features extracted during the analysis.</p>