Textual Abstraction: Latent Structure, Dimension Reduction
Barry deVille, Gurpreet Singh Bawa · 2021
One of the lively areas in text analytics is the area of topic modeling. This includes topic modeling techniques like latent semantic analysis, latent Dirichlet allocation as popularized by Blei, and the SAS approach to text topics described by Cox. These approaches employ a variety of statistical techniques to detect the underlying dimensionality in collections of textual data in order to infer the common topical content that is driving the observed behavior of the text. The dimensional products are formed by the matrix multiplication between the original matrix and the weights. This approach is aligned with a number of dimensional reduction approaches that include principal component analysis, factor analysis, and discriminant analysis. Latent semantic indexing is the process of establishing semantic content for a document based on its association with dimensional products, such as the singular value dimensions illustrated in this chapter.