Word2Vec-based Personal Trait Computing from User-generated Text

DOI 機関リポジトリ

この論文をさがす

抄録

type:Article

Personal trait is a habitual pattern for measuring behavior, thoughts, and emotions. It varies over individuals and is relatively stable in different situations over time. The personal trait is of great significance since it can be used in many applications, such as recommendation system, chatbot, and human resource management. Personal traits are easily recognized through wearable devices, social media, and the like. Most of the existing studies focus on user profile, behavior and personality. Specially, user profile and behavior are a person’s manifestations that cannot accurately capture a person’s internal characters. Personality is generally calculated by Big Five, which is obscure for non-psychologists. Generally, specific personal traits are especially critical in many aspects, such as disease detection, individual understanding, etc. Therefore, measuring more specific personal traits is essential. Given this, this paper proposes a word2vec-based general method for personal traits computing, which mainly includes topic word extraction, personal trait matrix generation, and personal trait computing. Furthermore, a case study is conducted to verify the effectiveness of the proposed method, and further analysis is provided to validate the methods.

収録刊行物

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ