Address structuring method and device

Abstract

The invention relates to an address structuring method and device. The address structuring method comprises the following steps that step 10, an address text is segmented into address word sequences; step 20, each address word in the address word sequences is subjected to part-of-speech tagging according to a predefined part-of-speech tagging set reflecting the address word features; step 30, the dependency parsing analysis is carried out on the tagged address word sequence according to the pre-defined address word dependency relationship rule, entity address words are used as nodes, the dependency relationship between the entity address words is used as edges, and a dependency parsing chart structure reflecting the address structure is generated. The invention also provides the address structuring device. The address structuring method and the address structuring device provided by the invention have the advantages that the dependency parsing chart structure can be efficiently and automatically generated for representing the dependency relationship between words in the address text, the manual intervention strategy is simple, and the knowing of a great amount of background knowledge is not needed; a dependency tree structure is expanded, so that the relationship between the address words can be expressed in a chart form.

Description

Technical field [0001] The present invention relates to the technical field of data processing, in particular to a method and device for address structuring. Background technique [0002] The system design research of a new generation of quantitative and qualitative fusion data management basic software with the goal of reliable, efficient, universal, and automatic processing of complex and distributed data includes how to capture meaningful matches in graph query and how to deal with the dynamics of graph data Research on characteristics and query complexity, how to query distributed graph data. [0003] As of 2012, the data set size unit that can technically be analyzed and processed in a reasonable time is exabytes. In many fields, due to the excessively large data sets, scientists often encounter limitations and obstacles in analysis and processing. For ordinary people, how to filter out the required knowledge in the face of large amounts of data has become more and more diff...

AI Technical Summary

Problems solved by technology

However, in the prior art, when analyzing natural language based on dependency syntax, the dependency tree is generally used to represent the dependency relationship, and the analytical dependency relationship of the natural language is mainly analyzed according to the linguistic grammatical characteristics such as subject, predicate, and object, and there is no address Special address structure research is carried out based on the characteristics of the data structure. At the same time, the simple tree structure cannot represent the complex relationship in the address.

Images