Entity information atlas generation method and device

Abstract

The invention discloses an entity information atlas generation method and device. The method comprises the following steps: collecting a text file; according to predefined class names and relation words, independently extracting a naming entity associated with each class name from each text file, and the attribute of the naming entity associated with each relation word; according to the attribute of the naming entity, independently carrying out association processing on the naming entities in each collected text file to obtain the entity relationship among all naming entities; according to a predefined event name, looking up the naming entity associated with the predefined event name, binding the predefined event name with the found naming entity; and taking the predefined event name as a clue, establishing mapping for associated information dispersed in each text file according to the extracted naming entity and the entity relation, and carrying out aggregation on the associated information to form an entity information atlas. The method can convert unstructured text data into structured data to realize multidimensional and complex knowledge atlas.

Description

technical field [0001] The present invention relates to natural language processing technology, in particular to a method for generating an entity information map and a related device. Background technique [0002] With the rapid development of the Internet, people are faced with an information explosion. Massive information is scattered on the Internet, which is fragmented, multilingual and international. The Internet is actually like a huge library. Every computer connected to the network is like a bookcase. This library has no catalog and is dynamic and rapidly increasing. At present, the work of the search engine is only to provide the position of the relevant books containing the keywords inquired by the user according to the keywords of the user, and find out the position of the books in the library. People are often submerged in the ocean of information. [0003] Due to the rapid development of the network, the dissemination speed of Internet information has increase...

AI Technical Summary

Problems solved by technology

[0006] 1. There are too many relevant information recalled by search engines, and it is difficult for users to locate the required information;

[0007] 2. Keyword-based search, although the matching algorithm is simple and easy to implement, it stays on the surface of the language without touching the semantics, and it is difficult to express the user's search intention clearly with the logical combination of several keywords;

[0008] 3. Even if the correct result is obtained, it is only a link to each independent article, which requires users to browse one by one;

[0009] 4. Unable to provide the physical and temporal correlation between articles, and reveal the internal connections and relationships of things

[0010] The development of the Internet has become a huge knowledge base, but because most of the information exists in unstructured data, people cannot organize and use this knowledge achievement, so there is no information and no knowledge.

Images