Name
Abstract | ||
|---|---|---|
This paper presents work that uses Transductive Latent Se- mantic Indexing (LSI) for text classification. In addition to relying on labeled training data, we improve classification ac- curacy by incorporating the set of test examples in the classi- fication process. Rather than performing LSI's singular value decomposition (SVD) process solely on the training data, we instead use an expanded term-by-document matrix that in- cludes both the labeled data as well as any available test ex- amples. We report the performance of LSI on data sets both with and without the inclusion of the test examples, and we show that tailoring the SVD process to the test examples can be even more useful than adding additional training data. The test set can be a useful tool to combat the possible inclusion of unrelated data in the original corpus. |
Year | Venue | Keywords |
|---|---|---|
2004 | FLAIRS Conference | singular value decomposition |
Field | DocType | Citations |
Training set,Transduction (machine learning),Singular value decomposition,Latent semantic indexing,Data set,Computer science,Matrix (mathematics),Artificial intelligence,Natural language processing,Labeled data,Machine learning,Test set | Conference | 13 |
PageRank | References | Authors |
0.78 | 12 | 1 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sarah Zelikovitz | 1 | 181 | 16.42 |