A CORPUS-BASED STYLOMETRIC ANALYSIS OF THE HARRY POTTER SERIES

Abstract

Text mining technique is mostly involved for the Stylometry of any literary or non-literary text. Previously, Stylometry technique is done manually to find out the written expression of various writers. Being reader it is a tiresome thing to read long texts and find out the words and distinction them as humanely not possible so, the purpose of this study is to extract out the J.K. Rowling’s written expressions through text mining technique using Voyant tool as it gives you authentic data (Zafar Ullah, Uzair, & Mahmood, 2019). Moreover, the major objective of this research to reveal the most frequent words of the writer to analyze and distinguish them. Firstly, data was generated as corpus and Summary panel was implied for the stylometric analysis on three alluring series of Harry Potter (1. Harry Potter and the Sorcerer’s Stone, 2. Harry Potter and the Chamber of Secrets, 3. Harry Potter and the Goblet of Fire as current study targeted stylistic characteristics qualitatively (interpretation) and quantitatively (statistical data). Mixed method approach was applied using Summary tool. Furthermore, the key-ground theory of Rakesh Aggrawal's Knowledge Discovery Theory (KDD) 1996 (Cabena, Hadjinian, Stadler, Verhees & Zanasi, 1998) which inquires about interesting and new knowledge patterns in dataset as it deals with Stylometry of texts. Major findings in all employs dialogic technique used with “said”. The vocabulary density 0.061 to 0.122 and the average length per sentence range12.3 to 14 which express simple and small sentences. The implications help educationist or novice readers to analyse different literal texts for making them concise.