UTS Statistika Lingkungan Virgiawan Malik Rizky

Lien du TP : https://www.mghassany.com/courses/ProbasBDML/exos3tp.html

This document presents a complete workflow for text mining and exploratory analysis using the crude dataset from the tm package in R. It is structured into two main parts: Data Loading and Preprocessing Loads the text corpus and applies standard cleaning steps such as lowercasing, punctuation and number removal, stopword elimination, and whitespace stripping. Summarizes the dataset with key statistics like document count, average length, and vocabulary size. Exploratory Data Analysis (EDA) Constructs a Document-Term Matrix (DTM) and computes term frequencies. Performs Unigram, Bigram, and Trigram analysis to identify frequent words and phrases. Visualizes results using bar plots and a word cloud for better interpretability. Highlights key observations about word distribution and contextual patterns. The analysis provides insights into the corpus structure and prepares the foundation for predictive text modeling using n-gram language models.

15 OKTOBER 2025 SEMESTER 5

ExamIDs2025B