rashitiwary / text-analytics-of-chapters Goto Github PK

Sentiment analysis

For this project, the dataset comprises an extensive document structured into 11 chapters, identified by Roman numerals (e.g., "I", "II"). To efficiently manage and process the text, a dictionary was created using the "which" function in R to organize the chapters. Once segregated, all the chapters were loaded into a VCorpus.

To enhance the text analytics process, I implemented several techniques, including the removal of stop words, the computation of Document-Term Matrix (DTM) and Term-Document Matrix (TDM), and the conversion of text into lowercase. Additionally, tokenization was applied to break down the text into individual units, and TF-IDF (Term Frequency-Inverse Document Frequency) was utilized to assign importance scores to words based on their frequency and rarity across the entire document. These text analytics concepts collectively contribute to a more refined and insightful analysis of the document content.