A high-order rule multi-classification method of knowledge system and its system

Abstract

The invention discloses a high-order rule multi-classification method and system of a knowledge system. The classification method comprises the steps that a participle text matrix is acquired; all object feature word texts belonging to the corresponding classification are acquired from texts in each classification of the participle text matrix, a one-order feature word matrix is constructed, and indexes between object names and notional words are set up; a two-order feature word pair text including object words and notional words at the same time is acquired from the one-order feature word matrix of each classification, a two-order feature word pair matrix is constructed, and indexes among the object names, the notional words and the classifications are set up; a three-order feature word group text including the object words, the notional words and field words is acquired from the two-order feature word matrix text of each classification, a three-order feature word group matrix is constructed, and indexes among the object names, the national words, the field words and the classifications are set up. A high-order classification model is set up through corpuses, accurate and fast classifying is achieved under the condition of large corpuses, the engineering implementation amount is large, and model modification is convenient.

Description

technical field [0001] The invention belongs to the technical field of information system classification, and in particular relates to a high-order rule multi-classification method of a knowledge system and a classification system thereof. Background technique [0002] With the improvement of the information system marked by search, people are increasingly caught in the difficult situation of information explosion, that is, among the large amount of searched information, people still need to spend a lot of time to read the contents of these items and select the real information from them. Required information entry. Therefore, the traditional information system can not meet the needs of people to acquire knowledge quickly. The method of traditional knowledge classification is the database method, that is, when information items are stored in the database, the attributes of the items, such as category, research object, author, author unit, etc., have been indicated. This is...

AI Technical Summary

Problems solved by technology

However, this method of manual collection, modeling, and verification will lead to technical problems such as a large amount of engineering implementation and uncontrollable classification results in the upstream research and development field with a large amount of corpus

Images