摘要：In the Internet world, the amount of information available reaches very high quotas. In order to find specific information, some tools were created that automatically scroll through the existing web pages and update their databases with the latest information on the Internet. In order to systematize the search and achieve a result in a concrete form, another step is needed for processing the information returned by the search engine and generating the response in a more organized form. Centralizing events of a certain type is useful first of all for creating a news service. Through this system we are pursuing a knowledge - events from the Internet documents - extraction system. The system will recognize events of a certain type (weather, sports, politics, text data mining, etc.) depending on how it will be trained (the concept it has in the dictionary). These events can be provided to the user, or it can also extract the context in which the event occurred, to indicate the initial form in which the event was embedded.
