Method and system for digital object classification

Abstract

The invention provides a method and a system for digital object classification. The method comprises the following steps of: obtaining a clustering method of digital objects; generating coarse classification methods of clustered sets, including a method for estimating classification parameters, so as to form a primary classifier; using the clustering result to regulate the parameter of the primary classifier, and determining a final classifier by being combined with a logical reasoning method. In one embodiment, the parameters are determined according to the primary classification result by adopting a likelihood estimation method, and the parameters are modified by adopting a posteriori estimation method of probabilistic reasoning so as to determine the final classifier, so that the influence of interference information is avoided effectively, and the defect of ambiguity caused by uncertain semantic information in the digital knowledge objects is remedied. By utilizing the classification method and the classification system provided in the invention, the accuracy and the expandability of classification of the digital knowledge objects can be improved.

Description

technical field [0001] The invention belongs to the field of knowledge management. It generally involves the classification organization, retrieval and mining of knowledge. Specifically, it involves automatically classifying and organizing computer-readable knowledge represented by digital objects through computer technology, and automatically providing the digital features necessary for retrieval and mining to the organized results. Background technique [0002] Currently, the rapid growth of knowledge available in the form of digital objects that can be processed by computers prevents people from fully understanding and effectively utilizing this large amount of information. How to help users organize this knowledge in an efficient way and find the key knowledge they need is a challenging task, and it is also the core purpose of the field of knowledge management. [0003] The learning of knowledge statistical relationship has become an important research hotspot in the f...

AI Technical Summary

Problems solved by technology

Each has a certain degree of defects: the supervised method needs a large number of training data sets in order to obtain the parameter estimation of the statistical relationship, which is difficult to obtain in a practical environment, especially in some fixed industry applications, and has poor applicability; the semi-supervised method will be affected by local The influence of the distribution of data features leads to the deviation of the overall parameter estimation. Although some studies have used the method of likelihood estimation to improve it, the effect is still not obvious when the computer automatically processes the process; the unsupervised method requires a strict pre-defined list of prior knowledge , such as keyword lists, etc., poor scalability

Images