Initialization in Gibbs Sampling Implementation of LDA

被引：0

作者：

Tekin, Yasar ^{[1
]}

机构：

[1] Techinsoft Bilisim Teknol AS, Merkez, Turkiye

来源：

32ND IEEE SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE, SIU 2024 | 2024年

关键词：

topic modeling; LDA; Gibbs sampling; random and fixed initialization;

D O I：

10.1109/SIU61531.2024.10600919

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Latent Dirichlet Allocation (LDA) is a text mining technique used for automatic extraction of topics addressed in document collections. Sampling-based or variational inference algorithms are used to approximate the posterior in LDA. In this study, the effect of random initialization in Gibbs Sampling implementation of LDA on model coherence is investigated. For this purpose, topic models using the same parameter values are initialized with random and fixed values, and model coherence scores obtained are compared with each other.

引用

页数：4

共 15 条

[1] Variational Inference: A Review for Statisticians [J].