Classification method based on machine learning knowledge

Abstract

The invention provides a classification method based on machine learning knowledge. The method comprises the steps of collecting classification behavior derivation classification reasoning rules of auser; obtaining knowledge to be classified; extracting a keyword word in the knowledge content by adopting a word segmentation method, the keyword word comprising a keyword and a word group; calculating the occurrence frequency and position of the keyword word; and screening the keyword words meeting the preset frequency and the preset positions, and outputting the keyword words as a classification result; extracting keyword words in the knowledge contents by adopting a word segmentation method, and searching the user classification behavior in the knowledge management platform; and establishing a nearest neighbor set according to the classification behavior of the user, and performing comprehensive analysis by combining keyword weighting, transformation, mixing, features and meta-level toobtain an optimally combined keyword, so that accuracy of label classification of knowledge is improved, retrieval accuracy is improved in the knowledge retrieval process, and information recommendation and information filtering are facilitated.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a classification method based on machine learning knowledge. Background technique [0002] The technique of grouping knowledge into pre-assigned categories dates back to the 1960s. However, in the last 10 years, due to the massive data brought by the digitization of text knowledge, we have to classify this knowledge. Therefore, the automatic classification of knowledge has been widely concerned and developed rapidly. [0003] At present, in order to solve the knowledge classification, there are mainly the following two methods: one is the method based on knowledge engineering, which defines a large number of inference rules for each category through professionals, and if the knowledge can satisfy these inference rules, it can be determined that it belongs to the category. category; the other is the knowledge classification based on the knowledge base, that is, f...

AI Technical Summary

Problems solved by technology

[0005] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a classification method based on machine learning knowledge, which is used to solve the problems of human participation, classification efficiency and accuracy when machines classify learning knowledge in the prior art. high problem

Images