Entity normalization processing method, device and equipment and storage medium

Abstract

The invention discloses an entity normalization processing method, device and equipment and a storage medium, and relates to an entity normalization processing technology. According to the specific implementation scheme, the method comprises the steps of receiving rule parameters related to an entity normalization strategy input by a user; generating a program code corresponding to the entity normalization strategy according to the rule parameter and a preset code generation rule; and running a program code corresponding to the entity normalization strategy, and performing normalization judgment on entities in a preset entity data set so as to cluster the same entities. The user only needs to input rule parameters related to an entity normalization strategy; program codes corresponding toentity normalization strategies can be automatically generated according to rule parameters and preset code generation rules. User programming is not needed, the manpower development cost and the learning cost are reduced, the threshold of data production is reduced, the entity normalization strategy is convenient to modify, the entity normalization processing efficiency is improved, and the method can be applied to entity normalization processing of data in any field.

Description

technical field [0001] The present application relates to the technical field of data processing, and specifically relates to entity normalization processing technology. Background technique [0002] In the construction of knowledge map data, since the construction of knowledge map often needs to use a variety of different data sources, it is an important task to normalize and fuse the same entities in different data sources. For example, the data of the movie "Children of the Weather" comes from three different websites, and its related attributes such as release time are 2019-11-01 (China), 2019-07-19 (Japan), 2019-11-01 (China) , the directors are Makoto Shinkai, etc., so they refer to the same entity, which needs to be disambiguated. The entity disambiguation process is divided into two steps: entity normalization and fusion. Entity normalization is to normalize the same entities into the same set; while in fusion, entities in the same set are fused, and strategies are ...

AI Technical Summary

Problems solved by technology

In the existing entity normalization method, R&D engineers need to program by themselves, which consumes a lot of labor costs, is difficult to learn, and lacks standardization guarantee; while using the model for entity normalization, the model training process requires a large amount of labeled data, and It requires professional algorithm engineers to iterate, it is difficult to apply to commercial scenarios, the industry is not universal, and it lacks applicability

Images