In terms of data, the tourism industry enjoys a significant advantage: A large majority of travelers post their journey online. From online searches for travel idea to posting photos on Flickr (by making hotel reservations via Booking), there is data everywhere. What role does artificial intelligence play in this sector, starting from marketing to travel?
In terms of data, tourism industry is complex to analyze. It has just 25% of structured data that comes from varied sources like business websites of tour operators, e-commerce by travel agencies, hosting companies, restaurants, transporters and CRM & other management software. The remaining 75% of unstructured data is scattered around the internet in forms of social media posts, photo albums, opinions and comments posted online. It is important to integrate metadata (age, sex, place of origin) with the textual analysis of all this data to add more context and value. Like all online marketing, the purpose of using this data is to increase sales. It, therefore, has the same limitations: too much personalization can be seen by buyers as intrusion and this can instead put them off.
Which tools help travel professionals maintain a balance between personalizing an offer and intruding into users’ privacy? The website Voyage Privé is determined to make up for the low user engagement and revisit rates on their site by improving the recommendations shown to users on the basis of their browsed images. In 2017, the company developed a tool that can analyze the images viewed by website users. It checks image properties like whether it contains a pool, a river or a beach and from this analysis, it automatically generates recommendations and even adapts images according to what the user is searching for. Behind this algorithm is a transfer learning technology developed with Dataiku using image databases (Places and Sun). The first layer of this neural network is available in open source. The site Usine-Digitale explains it as: ‘If a model is already capable of distinguishing a dog from cat in images, there is not much work to be done to make it capable of recognizing lions. This saves a lot of time […]. Voyage Privé has therefore reconstructed a deep learning model based on these databases and the first layers of a neural network that a third party had already built on Places database.’ With this innovation in testing phase, the tour operator has not yet measured the tool’s impact on sales.
Has internet buried the profession of travel agent or is it reinventing it? To expand in service sector via software, the IT giant IBM has been developing new artificial intelligence since 2005. Originally, IBM computer Watson was programmed to win the general knowledge based TV game show, Jeopardy where it won 3 times in a row in 2011. The IBM AI is today a group employing 2,000 people and co-incubating side projects like Wayblazer which is a tourism startup. Aimed at tourism professionals, this is a three-in-one platform: it interacts with users in their natural language, collects data to understand their profile and design appropriate recommendations and assists them throughout the purchasing process. Other AI applications are part of the ‘turn-key’ travel trend: from GPS based virtual reality tours to Google Trips app that suggests real-time tours, itineraries and restaurants from users’ emails and online searches, the concept of dematerialized travel concierge is becoming more of reality. These trends have given rise to more and more new applications for data-driven tourism. One such young start-up is Happs that uses machine learning to project ultra-personalized routes in interactive 3D maps of cities. All elements in these maps are clickable and informative.
In a 2014 report, the World Travel & Tourism Council had described two basic trends in our societies: denial of risk taking and overconsumption of information when it comes to travelling. And the surge of online data and information feeds these trends by offering innumerable risk analysis services, recommendations, opinions, itineraries, etc.
Can we say that the practice of travelling will gradually die? If it gets challenging in the face of expectations regarding safety of travelers and requisites for preservation of tourist sites, virtual reality offers some interesting and low-cost solutions. With 3D glasses, start-ups like Youvisit are already offering virtual tours of tourist sites. And to this, if haptic sensor technologies can also be added for physical sensations, the couch traveler will be fully satisfied.
Predicting the Solar Potential of Rooftops using Image Segmentation and Structured Data How could we automatically estimate the amount of electricity that solar panels could generate on any building of a country, instantly? It would require to be able to […]
How we improved our estimation of vegetated surfaces with Computer Vision Here at nam.R, we’re putting our hard work into building a Digital Twin of the French territory. We aggregate, clean and re-organize a large quantity of data of different […]
In the last few years, AI art has proliferated. Some AI algorithms can recreate an image in a painter’s style, some can improvise a track alongside a human musician on stage and some can even write a story presenting the […]
To detect all solar panels on the roofs of french buildings, we used aerial imagery and the known outlines of the buildings, through a pipeline consisting of a solar panel outline detector and a filtering algorithm. We took inspiration from […]
From September 24th to 25th 2018 Charles Hutin-Persillon, Data Strategist and Nicoles Berthelot, Lead Data Strategist, at nam.R participated to the french-norwegian seminar: “Observations, AI and sustainable development” “Observations, AI and sustainable development” Organised by the Ministry of Higher education, […]