A Tag Semantic Learning Method Based on Network Structure and Semantic Correlation Measure

Abstract

The invention relates to a label semantic learning method based on network structure and semantic correlation measurement, including initializing the label network based on user behavior facts to obtain a fact label network G; constructing a regulated label network G according to the fact label network G R ; in the label network G R Applying the improved random walk strategy to build a label network G based on the random walk strategy C ; Build a tag network G based on tag-related text information T ; for label network G C , label network G T Perform normalization processing, and learn label semantic vector representation through random walk strategy and word vector learning method. The invention has a reasonable design, not only makes full use of the network topology structure, but also takes into account the relevant text information contained in the nodes, and can learn from the topology structure and text expression to an easy-to-operate, high-confidence, full-expression, and low-noise system within a relatively short period of time. The tag semantic vector can be widely used in tag network learning and tag semantic learning of tagged text collections.

Description

technical field [0001] The invention belongs to the technical field of network representation learning, in particular to a tag semantic learning method based on network structure and semantic correlation measurement. Background technique [0002] In the network representation learning technology, text semantic learning is mainly to represent the features of the text, that is, to represent the target text object (word, sentence, paragraph, article) as a numerical value (single value, vector or matrix). In view of the needs of computing and semantic modeling applications, the currently commonly used models are Latent Semantic Analysis (LSA) based on matrix singular value decomposition, latent Dirichlet Allocation (LDA) based on probability model, and neural network-based solution The word vector representation model NNLM (Neural Network Language Model) and word2vec, etc., these models are mainly used for long text. In the face of short texts with high sparseness and high nois...

AI Technical Summary

Problems solved by technology

[0004] To sum up, the existing algorithms are difficult to solve the problems that the original label network comes from the unconstrained collaborative facts of users, the information is noisy and changeable, the topic boundary is not clear or the topic drifts, and the semantic model fails.

Images