Human–Machine Multi-Turn Language Dialogue Interaction Based on Deep Learning

Ke, Xianxin; Hu, Ping; Yang, Chenghao; Zhang, Renbao

doi:10.3390/mi13030355

Cited by 3 publications

(2 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In single-turn interactions, consumers express themselves either orally or manually, the technology responds by presenting output, and the interaction ends (e.g., Burggräf et al 2022; Ke et al 2022). For example, in a Google search, consumers enter search terms and the technology presents search results.…”

Section: Verbal Disclosure In Oral Versus Manual Interactions With Te...mentioning

confidence: 99%

“…In multiturn interactions, interlocutors engage in back-and-forth turn-taking, taking information received into account for the following turn (e.g., Burggräf et al 2022; Ke et al 2022). For instance, conversations with text or voice-based chatbots constitute instances of multiturn interactions (Bergner, Hildebrand, and Häubl 2019; Dellaert et al 2020; Luo et al 2019).…”

Section: Verbal Disclosure In Oral Versus Manual Interactions With Te...mentioning

confidence: 99%

See 1 more Smart Citation

Information Disclosure in the Era of Voice Technology

Melzner¹,

Bonezzi²,

Meyvis³

2023

Journal of Marketing

View full text Add to dashboard Cite

Whenever consumers interact with technological devices connected to the internet, they disclose information about themselves. The rapid diffusion of voice technology is shifting the way consumers interact with technological devices from typing or clicking to speaking. This article offers a comprehensive analysis of how this occurring shift from manual to oral communication with technology affects information disclosure. It first considers verbal disclosure and provides a conceptual framework that explicates how voice technology can influence consumers’ propensity to reveal information about themselves through semantic content disclosed voluntarily. It then considers non-verbal disclosure and provides an analysis of how voice technology enables the collection of information revealed rather unintentionally through vocal paralanguage and ambient sound. The article offers testable propositions and poses open research questions that can serve as impetus for future research. Additionally, it provides insights to marketers for how to navigate voice technology as a source of consumer information and to policymakers for how to better protect consumer privacy in interactions with voice technology.

show abstract

Section: Verbal Disclosure In Oral Versus Manual Interactions With Te...mentioning

confidence: 99%

Section: Verbal Disclosure In Oral Versus Manual Interactions With Te...mentioning

confidence: 99%

Information Disclosure in the Era of Voice Technology

Melzner¹,

Bonezzi²,

Meyvis³

2023

Journal of Marketing

View full text Add to dashboard Cite

show abstract

Research on Effective Information Extraction Techniques for Multi-Round Dialogues of Large-Scale Models in Deep Learning Environment

Hu,

Xiang,

et al. 2024

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

With the rise of deep learning technology, natural language processing applications have made significant progress, especially in the construction of large-scale model multi-round dialogue systems. The article proposes a multi-round dialogue intention recognition model based on the Transformer framework, which is applied to large-scale model multi-round dialogue and combined with the BERT-BiLSTM-CRF model to achieve effective extraction of multi-round dialogue information. The BERT model is used to obtain the relevant semantic vector information features of multi-round dialogues, and the BiLSTM model is used to annotate the sequences of multi-round dialogues, taking the sequential sequences of multi-round dialogues as the forward inputs and the reverse sequences as the backward inputs, so as to enhance the generation of temporal features of the dialogue information. The output of the BiLSTM model is then used as the input of the conditional random field, and the transfer characteristics between dialogue labels are fully considered to obtain the address annotation sequence with the largest joint probability to achieve effective extraction of dialogue information. To verify the feasibility of the model for effectively extracting multi-round dialogue information, simulations are carried out in this paper. The F1 value of the BERT-BiLSTM-CRF model for semantic extraction on the ATIS dataset is 96.09%, which is 3.65 percentage points higher than that of the BiLSTM-CRF model. As the number of iterations increases, the model’s loss value stably converges to 0.54 after the 10th iteration. Based on the BERT model, the combination of the BiLSTM model and the CRF model can achieve the effective extraction of semantic information from large-scale models for multi-round dialogues, which provides a new research direction for natural language processing.

show abstract

Can Gestural Filler Reduce User-Perceived Latency in Conversation with Digital Humans?

Kum

Lee

2022

Applied Sciences

View full text Add to dashboard Cite

The demand for a conversational system with digital humans has increased with the development of artificial intelligence. Latency can occur in such conversational systems because of natural language processing and network issues, which can deteriorate the user’s performance and the availability of the systems. There have been attempts to mitigate user-perceived latency by using conversational fillers in human–agent interaction and human–robot interaction. However, non-verbal cues, such as gestures, have received less attention in such attempts, despite their essential roles in communication. Therefore, we designed gestural fillers for the digital humans. This study examined the effects of whether the conversation type and gesture filler matched or not. We also compared the effects of the gestural fillers with conversational fillers. The results showed that the gestural fillers mitigate user-perceived latency and affect the willingness, impression, competence, and discomfort in conversations with digital humans.

show abstract

Human–Machine Multi-Turn Language Dialogue Interaction Based on Deep Learning

Cited by 3 publications

References 30 publications

Information Disclosure in the Era of Voice Technology

Information Disclosure in the Era of Voice Technology

Research on Effective Information Extraction Techniques for Multi-Round Dialogues of Large-Scale Models in Deep Learning Environment

Can Gestural Filler Reduce User-Perceived Latency in Conversation with Digital Humans?

Contact Info

Product

Resources

About