In this paper, we focus on performing LSI on very low SVD dimensions. The results show that there is a nearly linear surface in the local query region. Using low-dimensional LSI on local query region we can capture such a linear surface, obtain much better performance than VSM and come comparably to global LSI. The surprisingly small requirements of the SVD dimension resolve the computation res...

Latent Semantic Indexing (LSI) has proven to be a valuable analysis tool with a wide range of applications. however the crucial question, choosing an appropriate number of dimensions for LSI, is still unsolved. In this paper. a new method which is to deal with this problem is described. It finds that a sum of total dot products between all document vectors reaches the maximum value at a specifi...

In this paper we present a theoretical model for understanding the performance of Latent Semantic Indexing (LSI) search and retrieval applications. Many models for understanding LSI have been proposed. Ours is the first to study the values produced by LSI in the term dimension vectors. The framework presented here is based on term co-occurrence data. We show a strong correlation between second ...

In the area of information retrieval, the dimension of document vectors plays an important role. Firstly, with higher dimensions index structures suffer the “curse of dimensionality” and their efficiency rapidly decreases. Secondly, we may not use exact words when looking for a document, thus we miss some relevant documents. LSI (Latent Semantic Indexing) is a numerical method, which discovers ...

Singular value decomposition (SVD), the process at the heart of Latent Semantic Indexing (LSI), is a computationally expensive procedure. In this paper we analyze the relationship between higher order term cooccurrence and the values produced by the LSI process. We show a strong correlation between the number of cooccurrence paths and the value produced in the LSI term-term matrix.

This paper proposes the use of Latent Semantic Indexing (LSI) techniques, decomposed with semi-discrete matrix decomposition (SDD) method, for text categorization. The SDD algorithm is a recent solution to LSI, which can achieve similar performance at a much lower storage cost. In this paper, LSI is used for text categorization by constructing new features of category as combinations or transfo...

Abstract—We applied both Latent Semantic Indexing (LSI) and Essential Dimensions of LSI (EDLSI) to the 2010 TREC Legal Learning task. This year the Enron email collection was used and teams were given a list of relevant and a list of non-relevant documents for each of the eight test queries. In this article we focus on our attempts to incorporate machine learning into the LSI process. We show t...

Neisseria gonorrhoeae lipooligosaccharide (LOS) undergoes antigenic variation at a high rate, and this variation can be monitored by changes in a strain's ability to bind LOS-specific monoclonal antibodies. We report here the cloning and identification of a gene, lsi-2, that can mediate this variation. The DNA sequence of lsi-2 has been determined for N. gonorrhoeae 1291, a strain that expresse...

The analysis of speckle contrast in a time-integrated speckle pattern enables visualization of superficial blood flow in exposed vasculature, a method we call laser speckle imaging (LSI). With current methods, LSI does not enable visualization of subsurface or small vasculature, because of optical scattering by stationary structures. In this work we propose a new technique called photothermal L...