Tourism guidedance by AI - the best recommendation in town
Project Profile
Which tourist attraction can be recommended to a tourist as the next destination? We were on the trail of this question with the client. The next destination is certainly influenced by the previous POIs and may be reflected in the origin and many other parameters. We solved the matter with a recommendation engine that in any case creates a correct derivation as the next target from the previously learned targets per market. So we land the best recommender on the market which is absolutely independent from COVID!
To try out: https://ai.calista.at/ViennaInsights
At a glance - essential project data
| Duration | From 4/1/2020 to 7/30/2020 with about 4 months of full engagement | |
|---|---|---|
| Data and Tools | Market - Tourism Sources • Mobile data • Flight data VIE / BTL • Weather • Global vacation calendar | |
| Integration | Web API and Website for Demonstration Purposes https://ai.calista.at/ViennaInsights | |
| AI Methods | • Deep - Learning • LSTM Model with Embeddings |
Engagement Use-Case
Recommendation for an optimized visit itinerary (sequence of places or buildings) for tourists usable in CityApps embedding
Client motivation / Solution aims
Incentivize wayfinding for guides
tourists and tour groups
Design a maximum efficient offer for tourism programs individualized for all markets
To be able to recommend the ideal individualized itinerary for each touristic market in a city (Mobile Apps)
AI Approach
| AI key technology used in our solution | No comparable model known. Requirements were completely implemented in the neural network. | |
|---|---|---|
| Solution Approach | Multiple DeepLearning models (DNN) with embedding of origin countries and POIs as well as 3-stage memory. | |
| Project Approach | Simply agile | |
| Project Type | Proof-Of-Concept (POC) | |
| ML Integration and ML Operations | • Operation Integration API • Visualization: Website https://ai.calista.at/ViennaInsights |
Insights and Details
Small insight into the market behavior of tourists at POIs learned through embeddings.
The train through the city - redefined!
Which markets move through Vienna - which markets should be worked on together, because they have common ways - here the Miracle is solved
