Intelligent shopping assistant system

Wu, Bing-Fei; Yao, Shih-Jhe; Hou, Liwei; Chang, Po-Ju; Tseng, Wan Ju; Huang, Chun-Hsi; Chen, Yung-Shin; Yang, Po‐Yu

doi:10.1109/cacs.2016.7973916

Cited by 3 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the existing context-dependent text 80 generation systems are designed with the assumption of using template-based generators. An example of 81 text generation systems is the shopping assistant system (ISAS) that is able to plan a route within a 82 shopping mall for users who want to buy products according to their shopping list (Wu, et al, 2016).…”

Section: Context-dependent Generation Of Natural Language Texts 77mentioning

confidence: 99%

“…Although many computational message construction mechanisms were developed, most of them consider only static context information, e.g. weather, temperature, daytime or permanent things of a particular location, so as discussed in (Wu et al, 2016), (Lehsten et al, 2014) and (Braunhofer et al, 2013). However, the objective of this paper is to propose a real-time message construction mechanism (MCM) that generates both informative and instructive messages based on a real-time processing of dynamic context information.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Personalized Messaging Based on Dynamic Context Assessment: Application in An Informing Cyber-Physical System

Horváth

Rusák

2020

JID

View full text Add to dashboard Cite

Hazard-intense applications of cyber-physical systems (CPSs) such as the evacuation of a building on fire requires optimal management of stakeholders. Personalized message generation and informing of stakeholders based on real-time assessment of the dynamic context of stakeholders is the research and engineering challenge addressed by this paper. Personalized multi-message construction mechanism (MCM) that is enabled by dynamic context modeling, inferring and reasoning is proposed. Dynamic personal context was defined as the total of space-and time-varying situations that are relevant to a stakeholder. The basis of generating messages is a quantitative evaluation of the implications of the relevant situations with regards to the target stakeholders. The concept of impact indicator was used to represent the implications of situations and a personal danger level indicator was used to choose a proper message template for message construction. The algorithms included in the MCM were validated in a (simulated) indoor fire evacuation guiding application. Test people were involved in the practical evaluation of the quality of the generated messages. The conclusion is that the proposed MCM provides more sufficient information about personal context and expected actions than the messages constructed based on static context information.

show abstract

Section: Context-dependent Generation Of Natural Language Texts 77mentioning

confidence: 99%