Evaluation of background knowledge for latent semantic indexing classification

This paper presents work that evaluates background knowl-edge for use in improving accuracy for text classification using Latent Semantic Indexing (LSI). LSI’s singular value decomposition process can be performed on a combination of training data and background knowledge. Intuitively, the closer the background knowledge is to the classification task, the more helpful it will be in terms of creating a reduced space that will be effective in performing classification. Using a va-riety of data sets, we evaluate sets of background knowledge in terms of how close they are to training data, and in terms of how much they improve classification