Chapter 4: Discovering developmental trajectories and trends of conversational agent research using dynamic topic modeling
Restricted access

Conversational agents have been around for a few decades without going mainstream. However, in recent years, conversational agents have seen a surge in interest. At the most basic level, a conversational agent is a dialogue system built on artificial intelligence, enabling it to interact to a certain level of conversation with a human interlocutor through text or voice (Pérez, Daradoumis, & Puig, 2020). The first example of such, the chatbot ELIZA, was created by MIT professor Joseph Weizenbaum in 1966. Since then, conversational agents have been used in different fields for varying processes. The aim of this study is to reveal the developmental trajectories and current status of conversational agent research. The dynamic topic modeling method was used to analyze the diffusion of conversational agents technology in scientific literature. Dynamic topic modeling is a form of unsupervised machine learning to extract abstract topics from extensive collections of documents. The data for this study is exclusively acquired from the Web of Science (WoS) of Clarivate Analytics. The time span was restricted from 2001 to 2021, and the final search yielded 2,110 final documents. In the end, the topics that emerged from the data analysis were grouped under two main themes: usage areas and design and development.

You are not authenticated to view the full text of this chapter or article.

Access options

Get access to the full article by using one of the access options below.

Other access options

Redeem Token

Institutional Login

Log in with Open Athens, Shibboleth, or your institutional credentials

Login via Institutional Access

Personal login

Log in with your Elgar Online account

Login with your Elgar account
Edited by
Handbook