Conversation-Based Adaptive Relational Translation Method for Next POI Recommendation With Uncertain Check-Ins

Zhang, Lu; Sun, Zhi‐Wei; Zhang, Jie; Wu, Yiwen; Xia, Yunwen

doi:10.1109/tnnls.2022.3146443

Cited by 8 publications

(9 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The conversational recommender systems (CRSs) [1,41] has been becoming a hot topic in both industry and academy, due to the two significant advantages brought by conversations, viz., 1) enabling to dynamically capture the user's immediate preferences [43], and 2) revealing the reasons behind the recommendation for explainability. In this section, we first provide an overview of the related works from the angle of the data source that current CRSs are built upon, mainly categorized by real human dialogue and simulated dialogue.…”

Section: Related Workmentioning

confidence: 99%

“…Simulated Dialogue for CRSs. Another line of research on CRSs simulates dialogue for each user-item interaction record with the aid of side information, e.g., item features and knowledge graphs [29,43]. In particular, the conversation simulation is formulated as a slot-filling problem, where the asked features are filled in manually pre-defined templates.…”

Section: Related Workmentioning

confidence: 99%

“…Given the selected backbone datasets, we now simulate the dialogue between users and agents to obtain text-based conversations. Our study follows the simulated dialogue setting [43], wherein the CRS focuses on simulating such a scenario instead of real dialog. In this setting, the agent interacts with a user by asking for specific item features via slot filling [39] yet with fixed manual templates (e.g., Do you like track coat?).…”

Section: Text-based Conversation Generationmentioning

confidence: 99%

“…The natural advantages of CRSs, such as the ability to acquire user preferences through interactive conversations and provide intuitive interpretation for recommendations, align with the growing trend of explainable AI. However, the majority of research on CRSs to date [5,20,43] mainly focuses on text-based conversations between users and agents, also known as TCRSs, as illustrated in Figure 1. Nonetheless, TCRSs may not be as user-friendly or engaging for certain individuals, such as those with vision impairments or limited writing and reading abilities.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Towards Building Voice-based Conversational Recommender Systems: Datasets, Potential Solutions and Prospects

Qu¹,

Li²,

Sun

et al. 2023

Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

Conversational recommender systems (CRSs) have become crucial emerging research topics in the field of RSs, thanks to their natural advantages of explicitly acquiring user preferences via interactive conversations and revealing the reasons behind recommendations. However, the majority of current CRSs are text-based, which is less user-friendly and may pose challenges for certain users, such as those with visual impairments or limited writing and reading abilities. Therefore, for the first time, this paper investigates the potential of voice-based CRS (VCRSs) to revolutionize the way users interact with RSs in a natural, intuitive, convenient, and accessible fashion.To support such studies, we create two VCRSs benchmark datasets in the e-commerce and movie domains, after realizing the lack of such datasets through an exhaustive literature review. Specifically, we first empirically verify the benefits and necessity of creating such datasets. Thereafter, we convert the user-item interactions to text-based conversations through the ChatGPT-driven prompts for generating diverse and natural templates, and then synthesize the corresponding audios via the text-to-speech model. Meanwhile, a number of strategies are delicately designed to ensure the naturalness and high quality of voice conversations. On this basis, we further explore the potential solutions and point out possible directions to build end-to-end VCRSs by seamlessly extracting and integrating voice-based inputs, thus delivering performance-enhanced,

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Text-based Conversation Generationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards Building Voice-based Conversational Recommender Systems: Datasets, Potential Solutions and Prospects

Qu¹,

Li²,

Sun

et al. 2023

Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

show abstract

“…The work in this chapter is accepted by IEEE Transactions on Neural Networks and Learning Systems[100], 2022.…”

mentioning

confidence: 99%

Context-aware next point-of-interest recommendation with uncertain check-ins

Zhang¹

Self Cite

View full text Add to dashboard Cite

I would like to express my sincere gratitude towards my supervisor Prof. Jie Zhang, for his patient guidance, encouragement, and great support during my Ph.D. journey. His immense knowledge and insightful advice not only help me learn how to do research but also benefit my future career development. I feel very grateful to my co-author Dr. Zhu Sun for her constant support and elaborative guidance during the past three years. Many thanks for her shared generous experience in research. I would like to thank Prof. Gao Cong and Prof. Hong Xu for being my Thesis Advisory Committee members.I would also like to appreciate my group members for their friendly help. They include

show abstract