How to start using artificial intelligence
It seems that AI is very difficult, and in a small company
But it is not always necessary to compose something difficult.on their own. Big companies have already invented everything and put it in open access on GitHub. Both neural networks and smart libraries can be found there. For developers, this is a great opportunity to try new things and peek at how others solved the problem.
In Aviasales, many AI-related solutions are invented during internal hackathons.
Hackathon - this is a competition among developers, whenit is necessary to solve some problem in a very short time, for example, in 48 hours. Naturally, during this time it is impossible to create something from scratch, so ready-made solutions are used.
The most important thing - fast experiments with ready-made technologies almost always show good results, be it an increase in conversion or a reduction in costs.
"The Prophet" predicts when to buy a ticket
During one of the hackathons a hypothesis appeared that there is a connection between the time before departure and the day you start the trip. After analyzing a huge amountstructured data that has accumulated over the 11 years Aviasales service has been able to prove that the hypothesis is correct. This is how the Prophet service appeared, which predicts the best time to buy tickets with an error of 10%.
Thanks to the new service, the company began to saveon receiving third-party data and was able to substitute prices for those dates and directions in the calendar where real data was not available — with a small amount of error, the Prophet helps to find out the price in advance.
Travelers "The Prophet" gives tips aboutticket search time: “Buy Now” or “Wait.” Together with the words in the search, a chart is shown on how the price will behave based on the company's forecasts.
AI selects the best ticket seller
In the Aviasales meta-search, 200 ticket offices are presented.and 728 airlines. It is clear that in the first place is always the ticket with the lowest price. But a ticket can have several sellers, and often some have the same value. Then the question arises: who should be ahead?
The yellow button "Buy" - this is the first place among all sellers. Under the button is a list of agencies andairlines where you can also buy this ticket: for the same price or more expensive. To determine who to put on the magic button, two factors are taken into account - the commission that the partner pays for the ticket sold, and the conversion from going to the seller’s site to the purchase. That is, these are factors that take into account the interests of two parties - the meta-search and the convenience of the traveler.
All data on both factors are recorded inthe table. The data is constantly changing, as sellers are working to improve their sites. It was decided to automate this process in order not to enter the numbers into the table manually. So, in 5% of cases, on the “Buy” button, the seller turns out to be with not the lowest price to find out what percentage of users will go to his website and buy a ticket. Thus, the parameters are recalculated all the time, the system is trained on the basis of the obtained data and chooses the best solution itself.
AI chooses photos for the hotel description
If the choice of product or service is related to qualityphotos, and a huge number of them - manually select images irrational. Need an AI. The problem is that each partner sends its photos of the hotel to the Aviasales hotel service, and partners are not always network giants like Hilton or Marriott. Sometimes it is the owner of a small guesthouse in the Crimea, who photographed the rooms on the phone.
To analyze the photos, you need an AI, whichrecognizes quality and determines in which order to display images. The solution was found in one trained neural network that knows how to determine the location. The result is, for example, the following breakdown: 63% - building, 20% - swimming pool, 11% - wood, 6% - beach.
In city hotels it is interesting how the room lookstherefore, photos from the bed are shown first. In beach hotels, on the contrary, the pool and sun beds are important. As a rule, in resort areas the numbers are rather scarce, and the interior of the room is best shown last.
Starting to work with photos using AI,the company has reduced the cost of manual labor: previously hired freelancers who took pictures in popular cities and also increased conversion by 12%, mainly due to experiments with photographs of swimming pools at beach resorts.
How AI helps make beautiful websites with Weblium designer
Weblium website builder uses AI.Design Supervisor, which tracks user actions to create a site in real time and identifies design errors, correcting them on the fly.
Because the whole project uses a stack of productsGoogle, then for the implementation of this task, the developers used Google Cloud AI. The most difficult task was to teach the neural network to see design problems with incorrect use of colors, font pairs, and the like.
To get a large enough data set,The developers coached the model on a data set with 30 million design solutions taken from leading resources Behance and Dribbble. Site and item structures were recognized using the Cloud Vision API. This allowed us to make a "quantum leap" in achieving the accuracy of the work of AI Design Supervisor.
We can not yet boast that AI DesignSupervisor works flawlessly, but it can already be accurately used as the main point of differentiation from competitors. Users constantly write that turning one site into another on the fly invariably causes a wow effect even when AI Design Supervisor is used repeatedly.
David Brown, founder of Weblium
In parallel, Weblium is working tocontextually determine the content that the user brings to the sites, understand his tasks and offer him the most relevant blocks when building the site. For this, developers use the Cloud Natural Language API.
And the latest development, very important inperspective - voice interfaces. Weblium AI Lab prototypes the voice control of the site builder using the Cloud Speech-to-Text library. The final idea is that the user could put a technical task in a voice and in rather simple words, for example: “I want a modern functional site for my car wash”. And as a result of this TK, you get a decent site.
How Sephora and Lamoda AIs are used
Dynamic Yield personalization platform helps partners improve customer service. It is used by such well-known brands like Sephora and Lamoda.
Dynamic Yield can segment an audienceselect personalized products and content. The platform works on the web, on mobile devices, it can be used when sending newsletters and placing advertisements. Through all channels of communication, it delivers personalized recommendations to users.
Sephora has tested the personalrecommendations in eight online stores in Asia. In each of them, selected products were selected for users, guided by three strategies: similar products, related products, automatic recommendations.
Until the introduction of AI, the final choicegoods that will be shown to the user, was made depending on the country and KPI. Now they are displayed depending on which products the user added to the basket and which ones he bought in the end.
Thanks to this approach, the CTR increased by 4%. And every dollar spent on use, Dynamic Yield earned $ 6.5 in revenue.
Lamoda has previously segmented users bylocation and recommended clothing appropriate to weather conditions. Now the recommendations are based not only on geo, but also on the history of purchases, user behavior, preferred brands and products.
Lamoda divided users by 160microsegments and prepared personalized coupons for each segment. Compared with the previous discount campaign, this has increased conversion, the average income per visitor and revenue per session.
Lamoda launched personal banners on whichdisplayed clothes and accessories of the color that the user preferred during the search. When clicking on a banner, the user saw clothes of his favorite color, sorted in the order that he usually prefers when searching.
Through the use of AI, Lamoda increased revenue by session by 8%, and gross profit rose $ 15 million.
Ready solutions, quick effect.
Examples Aviasales, Weblium, Sephora and LamodaProve that the use of artificial intelligence helps companies grow significantly, and sometimes in a short time: from several months to a year. Moreover, some indicators could never be improved without introducing AI.
You can start experimenting with AIquickly. At the initial stage, most likely, there will be enough power of those developers that you already have. Look at GitHub for development that can be adapted for your company, find out if the use of a completely finished third-party product will pay off, and try to implement at least a small idea to look at the results. Surely they will impress you so much that you will continue to experiment with AI.