Method and device for adding foreign language business object category labels

Abstract

The invention discloses a method and device for adding foreign language business object category labels. The method comprises the steps that relevant data are extracted from a preset foreign language business object training set; input second business object titles of foreign language business objects to be classified are received; according to first business object titles, business object categories and the second business object titles, the probability values of the foreign language business objects to be classified under business object classes are calculated; the two maximum probability values in the probability values obtained through calculation are obtained; when at least one of the two maximum probability values is larger than a first threshold value, a first specific value of the two maximum probability values is calculated; when the first specific value is larger than a second threshold value, category labels corresponding to the business object classes with the maximum probability value in the two maximum probability values are added to the foreign language business objects to be classified. According to the method, the accuracy for adding the foreign language business object category labels can be ensued, and the efficiency for adding the category labels can be improved.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method for adding a category label of a business object in a foreign language and a device for adding a category label of a business object in a foreign language. Background technique [0002] In an e-commerce website, it is often necessary to classify a large number of business objects and add category labels to clearly classified business objects. [0003] For most users, due to language barriers in foreign language business objects, how to ensure accurate classification of foreign language business objects and add category tags is an urgent problem to be solved by every e-commerce website. [0004] In the prior art, the common practice is to separately calculate the probability values ​​of the foreign language business objects to be classified under each business object classification, and then obtain the business object classification corresponding to the maximum probability ...

AI Technical Summary

Problems solved by technology

[0005] However, when the calculated probability values ​​under the classification of each business object are not much different, the practice of the prior art is likely to lead to the problem of wrongly adding category labels, and then it i

Images