Research on Topic Evolution Path Recognition Based on LDA2vec Symmetry Model

Topic extraction and evolution analysis became a research hotspot in the academic community due to its ability reveal development trend of certain field discover law topic content different stages field. However, current methods still face challenges, such as inaccurate recognition unclear paths, which can seriously compromise comprehensiveness accuracy analysis. To address problem, paper proposes path method based on LDA2vec symmetry model. Under given conditions, both LDA Word2vec used model conform structural their datasets high-dimensional space, fused improves results. Firstly, we recognize topics model, uses Gibbs symmetric sampling obeys Dirichlet distribution ensure data convergence. Secondly, is learn contextual information words document collection, corpus are projected vectors space so that computed pairs with similar semantics have hyperplane space. Subsequently, word vector weight, probability value weighted generate new vector. Thirdly, similarity index employed calculate semantic among at adjacent stages, paths directly reflect relationships constructed. Finally, an empirical study conducted security demonstrate effectiveness proposed approach for The results show accurately construct clear contribute comprehensive accurate specific