A Virtual Assistant for Natural Interactions in Museums

Abstract

Artificial Intelligence (AI) and its real-life applications are among the most effervescent research topics of the last couple of years. In the past decade, stakeholders such as private companies, public institutions, non-profit entities, and even individuals, have developed and used various AI algorithms to solve a wide range of problems. Given the extended applicability and the disruption potential of this technology, it was just a matter of time until it connected to the field of cultural heritage. This paper presents the development of an intelligent conversational agent which was built to improve the accessibility to information inside a history museum. We present the cultural context, the application architecture, the implementation challenges, and the visitors’ feedback. We created a smart virtual agent that interacts with users in natural spoken language. This involved the design and implementation of the artificial intelligence-based core responsible for understanding the Romanian language. A short survey regarding the tourist acceptance of the system was conducted at the premises of our partners, the Museum “Casa Mureșenilor” from Brașov, shows good acceptance levels from both visitors and museum staff. Given the flexibility of the implementation, the system can be used by a large array of stakeholders with minor modifications.

Stand by me: analyzing the tourist–intelligent voice assistant relationship quality

“…robots with human traits) as a way to go further in the social interactions and the relationships between humans and IVAs. The gender of the voice simulated in IVA should also deserve further investigation, as it can affect relationships among humans (Duguleană et al , 2020). Future studies can also devote their attention to the influence that information society’s inequality – known as digital division – can have on the relations mentioned above (Van Dijk, 2005).…”

“…Recent studies explore voice assistants either in a more general context (Zhao and Rau, 2020) or in more specific contexts such as advertising (Lee and Cho, 2020) or education (Beneteau et al , 2020). In the hospitality context, some studies devote attention to text chatbots (Pillai and Sivathanu, 2020; Shin and Jeong, 2020) and few studies focus their attention on the interaction between IVA and humans, which is critical for hospitality organizations (Duguleană et al , 2020; McLean et al , 2020). The travel and hospitality industry is mainly associated with hedonic services, where sensory benefits such as playfulness, fun and pleasure for tourists and guests, are key (Li et al , 2019) and AI-embedded technology is becoming more relevant for communicating and relating with tourists (Belanche et al , 2019).…”

Closing the Gap between Museums and Schools

“…These programs enhance the museum experience of school students and facilitate direct and live contact with a museum's heritage and its specialists, not only its educators. AI offers additional opportunities for museums to find new, entertaining, and engaging ways to reach their school audiences (and not only them) (see some general consideration related to the adoption of AI in museums in Charr, 2020Charr, , 2021Duguleană et al, 2020;Murphy & Villaespesa, 2020;Vidu, Zbuchea, & Pinzaru, 2021; and specific approaches related to museum education in Aslan, 2021).…”

Online Museums Segmentation with Structured Data: The Case of the Canary Island’s Online Marketplace

“…With this in mind, lead generation allows us to know how the customer perceives certain content and how they interact with it, being a clear demonstrator of their level of satisfaction and relationship with the brand. Thus, efforts are focused on analysing and identifying the existence of some data collection mechanisms such as using interactive story telling techniques [74,75] and generating leads such as subscription newsletters [54], form completions [76,77] or the possibility of getting opinions and new ideas from users through the existence of the complaint and claim forms [78] as well as by managing community platforms [79]. On this basis, Hypothesis 7 is put forward as follows: Hypothesis 7.…”