Inferring gender of movie reviewers: exploiting writing style, content and metadata

摘要: Despite differences in the way that men and women experience goods communicate their perspectives, online review communities typically do not provide participants' gender. We propose to infer author gender, given a set of reviews particular item, experiment on posted at Internet Movie Database (IMDb). Using logistic regression, we explore contribution three types information: 1) style, 2) content, 3) metadata (e.g. age, social feedback). Our results concur with previous research, there are salient writing style content between authored by versus women. However, comparison literary or scientific texts, which classification tasks often applied, brief occur within context an ongoing discourse. Therefore, compensative for brevity reviews, stylistic features can be augmented metadata. find perceived utility is important correlate The model incorporating all has accuracy 73.7% as sensitive length those based only features.