2022
DOI: 10.1109/tcds.2021.3086565
|View full text |Cite
|
Sign up to set email alerts
|

Dialogue Management in Conversational Systems: A Review of Approaches, Challenges, and Opportunities

Abstract: Attracted by their easy-to-use interfaces and captivating benefits, conversational systems have been widely embraced by many individuals and organizations as side-by-side digital co-workers. They enable the understanding of user needs, expressed in natural language, and on fulfilling such needs by invoking the appropriate backend services (e.g., APIs). Controlling the conversation flow, known as Dialogue Management, is one of the essential tasks in conversational systems and the key to its success and adoption… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
8
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
2
2

Relationship

2
7

Authors

Journals

citations
Cited by 21 publications
(8 citation statements)
references
References 97 publications
0
8
0
Order By: Relevance
“…Agents typically use pattern matching and string processing, but more advanced technologies enable agents to use complex knowledge-based models (Hussain et al, 2019), and emerging technologies will enable multiple agents with different strategies and communication styles at the same time (Lippert, Shubeck, Morgan, Hampton, & Graesser, 2019). Despite the advancements, effective, scalable, and robust dialog management techniques are still a challenge, but, for example, explainable AI could make a dialog management system more human-like as users can understand the rationale behind it (Brabra et al, 2021). However, at least for now, PAs are rather responsive to students' communication actions and cannot maintain a conversation, and, therefore, PA interaction cannot be called a reciprocal communication process.…”
Section: Discussionmentioning
confidence: 99%
“…Agents typically use pattern matching and string processing, but more advanced technologies enable agents to use complex knowledge-based models (Hussain et al, 2019), and emerging technologies will enable multiple agents with different strategies and communication styles at the same time (Lippert, Shubeck, Morgan, Hampton, & Graesser, 2019). Despite the advancements, effective, scalable, and robust dialog management techniques are still a challenge, but, for example, explainable AI could make a dialog management system more human-like as users can understand the rationale behind it (Brabra et al, 2021). However, at least for now, PAs are rather responsive to students' communication actions and cannot maintain a conversation, and, therefore, PA interaction cannot be called a reciprocal communication process.…”
Section: Discussionmentioning
confidence: 99%
“…The 5th article [16] does not focus on any specific virtual assistant or conversational system, but rather provides a broad overview of different approaches and challenges in dialogue management. Therefore, it does not describe a specific method used by a virtual assistant to understand user input.…”
Section: Sourcesmentioning
confidence: 99%
“…Dialogue systems in roleplaying video games are task-oriented (Grosz, 1974), where quests in the game act as tasks that the player must complete within the constraints of the game world. The static pre-written dialogue graphs are a form of finitestate dialogue management (Brabra et al, 2022). Using such a rigidly constrained dialogue management approach ensures the authorial intent of the game writers at the expense of more natural conversation flows.…”
Section: Related Workmentioning
confidence: 99%