A Stylometric Approach to Agatha Christie’s Works Using Random Forests : Comparing with Dorothy Sayers

Random　Forests　ヲ　シヨウ　シタ　Agatha　Christie　サクヒン　ノ　ブンタイ　ブンセキ　Dorothy　Sayers　ト　ヒカク　シテ

ツチムラ, ナルミ

Tsuchimura, Narumi

Graduate School of Language and Culture, Osaka University

オオサカダイガク　ダイガクイン　ゲンゴブンカ　ケンキュウカ

This study tries to make a quantitative analysis on characteristic words in Agatha Christie works, comparing with her contemporary Dorothy Sayers. This study applies Random Forests, one of ensemble learn in g method s, for classifying the two writers ’works. The variables used in Random Forests are the most frequent words, and we discuss the 100 the most characteristic words. The results of this analysis revealed that Christie’s use of characterist ic words contrasted with Sayers’，especially between synonyms: Christie tends to use someone, anyone and until while Sayers tends to use somebody, anybody and till. Moreover, Christie tends to use words related to the female gender and a number of past tense verbs. Among the past tense verbs, we focus on the words related to movement. It revealed that Christied escribed characters' behaviour in her novels by using a number of co-occurring words like-ly adverbs, and I assume that this makes readers to imagine scenes in her novels easily. It may be one of reasons why her novels are read and loved all over the world now.