2022/10/04 by Marijn Koolen, Julia Neugarten, Peter Boot
#Goodreads #impact model #online book reviews #reading impact
paper · doi:10.48694/jcls.104
Online book reviews are an important source of data for analysing how people read books and how books influence them. Being able to identify and analyse reading impact expressed in reviews allows us to investigate how books affect their readers. In this paper we investigate the feasibility of creating an English translation of a rulebased reading impact model for Dutch book reviews. We extend the model with additional rules and categories to measure reading impact in terms of positive and negative feeling, narrative and stylistic impact, humor, surprise, attention and reflection. We created ground truth annotations to evaluate the model and find that the translated rules and new impact categories are effective in identifying reading impact expressed in English book reviews. Additional rules are needed to improve recall and some impact aspects are hard to extract with our type of rules. When applying the model to a large set of reviews, lists of the top-scoring books in the categories show the model’s prima-facie validity. Correlations among the categories include some that make sense and others that require further research. Overall, the evidence suggests this is a suitable approach for investigating the impact of books.