Companies, especially those claiming to be technology companies, cannot afford to spend too long looking at innovation. Therefore, the implementation of AI/ML technologies has turned into a kind of competition: whoever was the first to understand exactly how this will help speed up or optimize work, reduce costs and improve user experience will be the leader. Therefore, for companies engaged in online trading and providing services, such as classifieds, marketplaces or banks, the introduction of artificial intelligence is already a top priority. (View Highlight)
Artificial intelligence has been called the “universal equalizer.” Judging by research, it gives stronger employees a 15% increase in productivity, and weaker ones - 30%. The same thing happens with companies. Large and successful market leaders and new startups now seem to be equal in capabilities, which means that competition has instantly intensified, and no one can simply rest on their laurels. (View Highlight)
AI/ML technologies give to classifieds to significantly improve their work? One use case is filling out announcement cards. Such solutions are already being tested, for example, on the AliExpress marketplace, where artificial intelligence, when the seller fills out a product card, offers the best options for title and description. For people putting their apartment up for sale, creating a clear description is not always easy. Artificial intelligence can help with this or even take over this work entirely. In the future, it will probably be enough for Qian users to enter the address of their property and upload photos; the AI will do the rest itself, highlighting the main advantages and parameters for buyers. (View Highlight)
The capabilities of machine learning and artificial intelligence algorithms can also be useful to realtors. For example, with the same filling of ads, of which they have to create a lot more. Optimizing the process will allow you to get rid of routine, and not only that associated with writing advertisements. The lion’s share of a realtor’s time is occupied by calls, conversations, arrangements for viewings, and planning meetings. An AI-based assistant can do all this instead: answer typical questions about the apartment, find free time through the calendar to organize a viewing, and notify the property owner and potential buyer about this. And, of course, all or almost all work with documents. Formation of an agreement, sending requests to the bank or all documentation for registering a transaction - all this can be delegated to AI. (View Highlight)
Creating a contract, like any other document, no longer requires special human participation if we are talking about a template. But each apartment, as a rule, has many “bottlenecks” that must be taken into account in the contract: heirs, lack of ownership documents, debts or rights of third parties. Such subtle nuances go beyond a simple template, and to correctly display them in the contract and in the transaction as a whole, you can involve not only a lawyer, but also solve them using artificial intelligence. (View Highlight)
The introduction of artificial intelligence into the work of Classified can help relieve not only the pain of users, but also the platforms themselves. User safety and protecting them from scammers is one of our priorities. The use of predictive models that take into account account history and user behavior patterns allows you to identify fraudsters much faster and more efficiently than manually. Now this happens already in the first minutes of appearance on the site, that is, even before the first contact with users is made. In this way, we minimize the risks from fraudsters and provide a safe user journey. (View Highlight)
Also, one of the key problems for all advertisement portals has been unreliable advertisements, the so-called “lanterns,” for many years. They are used for lead generation. When a call is received regarding such advertisements with the most attractive conditions, a good property and a price below the market, the realtor informs the buyer that the property has already been purchased and redirects him to an actual apartment. Using AI, we have learned to automatically identify 80% of unreliable ads before they are reported. To identify such objects, we analyze the correspondence of the description to the entered address, photographs of the object and ad parameters to the description; we can convert call recordings into text and use a test model to catch suspicious signals. Thanks to the work of machine learning and artificial intelligence algorithms, we are able to ensure high quality content and minimize the number of unreliable objects. According to this indicator, we are 30% better than other classified companies, and in the capital regions - twice as good. (View Highlight)
The introduction of systems built on the basis of artificial intelligence will allow for automatic moderation of ads. In the future, it can evaluate the ad’s compliance with reality. The use of AI in such processes can significantly reduce operating costs: in our processes we are talking about tens of millions per year. (View Highlight)
New highlights added May 17, 2024 at 5:23 PM
For many years, companies have strived to use knowledge about their customers to create the most personalized offer possible. In the industry of online platforms, including classifieds, it is now generally accepted that the next stage of our development is precisely the maximum individualization of interaction with the client. We believe that the introduction of machine learning and artificial intelligence algorithms will allow us to create a personalized offer based on previous user experiences, including search queries, browsing history, and so on. (View Highlight)
We believe that thanks to AI in the near future it will be possible to form search results in accordance with the real needs of a particular person, and not, as now, formally according to parameters. Already now, classification search engines display offers and recommend the most suitable ones to the user. But the use of AI allows us to make this offer better and more relevant to current needs. Intelligence will take into account what ads a person reacted to, how he changed the wording of the request, and what he wrote in chatbots. (View Highlight)
On the way to changing the search format, we were the first among Russian real estate classifieds to launch a smart assistant - CianGPT. He independently searches for suitable real estate of all types using specified text parameters, and also knows how to answer simple questions that are usually answered by the support service. (View Highlight)
Thanks to artificial intelligence and machine learning algorithms, CianGPT personalizes the search and sends selections to the user based on his text request. For example, a request might look like this: “Send me one-room apartments in Nizhny Novgorod from 5 to 10 million rubles” or “Find apartments for rent in Moscow in the North up to 50,000 rubles.” The criteria can be more or less detailed - based on this, the assistant will send options with a certain frequency, which you can also set yourself by clicking the commands “Come again”, “Stop”, “Set up frequency” and so on. The response speed takes only a few seconds. In addition to searching for real estate, the assistant also answers frequently asked questions: “How to get a mortgage” or “How to sell an apartment” and others. (View Highlight)
We see that in foreign classifieds, fresh solutions based on a chatbot are also increasingly appearing, which interactively asks the question: “what are you looking for?”, “Where are you looking for?”, and the user answers him. The task of the chatbot is to compare these requests with the existing database of objects and select from it those that fit the stated parameters. The CianGPT assistant works the same way. In the future, such a search system will take into account not only the request itself, it will “remember” whether the person has a family, what stores he was interested in, whether he drives a car and whether he goes to the movies. Thanks to this, the proposal will hit the target as much as possible. (View Highlight)
The “smart” classification of the future will have to guess what the user wants, and not only based on what he was “told” in the search bar. Nowadays, the output consists of thousands of objects, and the user simply does not have enough time to view them all. People can spend months looking for an apartment, and no one likes this state of affairs. In the future, the use of AI should relieve users of the need to choose from an entire array of objects, 99% of which are obviously not suitable for them. Therefore, searching for real estate will never be the same. (View Highlight)