Grants for projects

AIACE

Artificial Intelligence Assistant for Cultural Tourism Enhancement
Digital narratives
Digital tourism
AIACE
Scheda del progetto
Responsabile
P.i Scientific Resp.
Prof. Edoardo Tortarolo
Università del Piemonte Orientale
Team
Team
Prof. Edoardo Tortalo
Prof. Davide Porporato
Prof. Vito Rubino
Dott.ssa Irene Gaddo
Data
Start date
Durata
Duration
8 months
Valore
Approved value
€ xx
Investimento
Investment nodes
€ xx
Dipartimento
Department
Università del Piemonte Orientale

The so-called LLMs (Large Language Models) are based on generic contents and, although extremely broad, lack precision, giving rise to errors or "hallucinations" in a non-negligible number of cases, varying from 3% to 27% of the answers based on the different models. The AIACE project is specific: it is intended to pave the way for the creation of AI tools that function as a support for the tourist use of territories and learning in the humanities, while models such as ChatGPT are generic. The expected results consist in demonstrating the effectiveness of this approach as a support in the dissemination of knowledge in the historical and more generally humanistic fields.

 

 

Contacts:

Ref.
Prof. Edoardo Tortalo

Email

edoardo.tortarolo@uniupo.it

 

 

The challenge
Document

By querying the AI ​​via the web, you will obtain responses in natural language whose accuracy and breadth will exceed the current offer of information, making the user autonomous and/or supporting the work of guides, museum operators, etc. in the territories concerned. The Proof of Concept is also proposed as a model that can be replicated in other local areas, with potential implications that make it attractive in terms of technological transfer from research to the market.

Because it is innovative
Document

The innovative potential of the AIACE artificial intelligence assistant lies in its ability to offer users a tourist experience enriched by dialogic interactions. supporting him in obtaining historical-cultural information in the tourism sector. In response to questions provided as input by the user, the AI ​​assistant is able to develop a series of relevant answers, aimed at immediately resolving the problem.

Impact on those who use it
Document

Advantages   Applications:


• Scientific validation of the knowledge base used for training
• Easily extendable to different disciplinary domains and cultural fields
• Centered on Italian and therefore free from the linguistic biases of the models
  trained on English texts