Name
Abstract | ||
|---|---|---|
Latent Dirichlet Allocation (LDA) is a probabilistic framework by which we may assume each word carries probability distribution to each topic and a topic carries a distribution to each document. By putting all the documents together into one collection by each author, it is possible to identify authors. Here we show that author identification is fully reliable within a framework of LDA independent of documents domains by learning incomplete and massive documents. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1007/978-3-642-23878-9_2 | IDEAL |
Keywords | Field | DocType |
probabilistic framework,documents domain,latent dirichlet allocation,massive document,domain independence,author identification,probability distribution,text mining | Latent Dirichlet allocation,Domain independence,Information retrieval,Computer science,Probability distribution,Probabilistic framework | Conference |
Volume | ISSN | Citations |
6936 | 0302-9743 | 4 |
PageRank | References | Authors |
0.51 | 4 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Masato Shirai | 1 | 4 | 1.52 |
| Takao Miura | 2 | 61 | 22.10 |