Search engines use thematic search as the main one - that is, they give out links based on the words included in the request. This option, for all its convenience, does not have "intelligence" - that is, the search engine does not understand what exactly the user is looking for, the search is carried out mechanically by the coincidence of words. The situation can be changed by semantic search, which can qualitatively improve search results.
Instructions
Step 1
Modern subject search is good at dealing with a situation where the user knows exactly what he is looking for and makes the right search query. Nevertheless, the search results always contain a lot of unnecessary links, at best, indirectly related to the subject of the search.
Step 2
An alternative to classical search can be semantic, the algorithm of which is built in such a way that the meaning of words in the search query is taken into account. In this case, the user receives not only information about the sites where these words were mentioned, but also specific information corresponding to the essence of the search query.
Step 3
For example, if he enters a request to observe the Moon, then the user will receive information about the history of studying and observing the Moon, about the observation technique, and the necessary equipment. In the most complete version, the user's location (by IP-address) can be taken into account, therefore, relevant information will be given about the most convenient time of day for observing the Moon in the user's area.
Step 4
Thus, when using semantic search, the system itself collects and provides the user with the information he needs, and does not send him to the set of found resources. At the same time, the links themselves will not disappear anywhere, the user will always have the opportunity to view them.
Step 5
Many search services have tried and are trying to implement semantic search. For example, Powerset has provided a convenient way to search one of the most popular online sources of information - Wikipedia. However, this solution is far from perfect, as in many cases the information from Wikipedia is not the best. The search service of the Hakia company, which collects information from the entire Internet, has been improved.
Step 6
Google introduced its version of semantic search with the launch of the Knowledge Graph service. When you enter a query into the search bar, Google correlates it with a database that contains about half a billion objects. As a result, the user receives a search result in the right block that already contains the bulk of the necessary information. At the moment, the service is only available in English.
