Dynamic Topic Model for Quantitative Estimation of Conversation Process During Psychotherapy

DOI Web Site
  • Yokoyama Satoshi
    Graduate School of Biomedical and Health Sciences, Hiroshima University
  • Takagaki Koki
    Health Service Center, Hiroshima University
  • Kambara Kohei
    Graduate School of Humanities and Social Sciences, Hiroshima University
  • Jinnin Ran
    Graduate School of Biomedical and Health Sciences, Hiroshima University
  • Okamoto Yasumasa
    Graduate School of Biomedical and Health Sciences, Hiroshima University

Bibliographic Information

Other Title
  • 動的トピックモデルを用いた心理療法における会話プロセスの量的推定
  • ドウテキ トピックモデル オ モチイタ シンリ リョウホウ ニ オケル カイワ プロセス ノ リョウテキ スイテイ

Search this article

Abstract

<p>Limited evidence has shown how the therapeutic effects of psychotherapies have been obtained compared to abundant evidence based on the changes in the clinical outcome scores. Psychotherapy depends on interactive therapeutic conversations. Hence, topic models have recently attracted attention for linguistic information as data-driven computational approaches can produce objective and reproducible models. This study examined whether a dynamic topic model with a time-series structure could extract treatment session content from single-case conversation data during a structured behavioral activation program. We extracted four potential topics during the treatment sessions. The quantitative transition process reflected the program treatment contents. This is the first study to apply the dynamic topic model for conversation during psychotherapy, which quantifies the actual treatment process. The results show cumulative evidence of the effects of psychotherapy.</p>

Journal

Related Projects

See more

Keywords

Details 詳細情報について

Report a problem

Back to top