The Best-Selling Algorithm

A group of scientists at Stony Brook University have developed an algorithm which can predict with 84 per cent accuracy whether a book will be a commercial success. Using a technique called “statistical stylometry,” they examine an author’s use of words and grammar. According to their study, successful books tended to include heavy use of conjunctions, and used a large number of nouns and adjectives, focusing less on verbs and adverbs. Furthermore, books that favored verbs describing thought-processes (e.g. “recognized” and “remembered”) and verbs that served to quote and report (e.g. “say”) were more successful than books that relied on verbs that are explicitly descriptive of actions and emotions (e.g. “promised,” “cried,” and “cheered”). The study drew on a number of works available through Project Gutenberg, focusing on first works from previously unseen authors. The scientists write:

“Predicting the success of novels is a curious ques-
tion among publishers, professional book reviewers,
aspiring and even expert writers alike. There are po-
tentially many influencing factors, some of which
concern the intrinsic content and quality of the book,
such as interestingness, novelty, style of writing, and
engaging storyline, but external factors such as so-
cial context and even luck can play a role. As a re-
sult, recognizing successful literary work is a hard
task even for experts working in the publication in-

For those who are looking for a little writing advice outside of the ubiquitous Strunk & White, this article might give you a thing or two to think about. And for those interested in natural language processing, the article is a must-read. See it here:

Ashok, Vikas Ganjigunte, Song Feng, and Yejin Choi. “Success with Style: Using Writing Style to Predict the Success of Novels,” Emperical Methods in Natural Language Processing (EMNLP), 2013.

Twitter Email