In this paper, we propose an acceleration of collapsed variational Bayesian (CVB) inference for latent Dirichlet allocation (LDA) by using Nvidia CUDA compatible devices. While LDA is an efficient Bayesian multi-topic document model, it requires complicated computations for parameter estimation in comparison with other simpler document models, e.g. probabilistic latent semantic indexing, etc. T...

Learning Plausible Verb Arguments allows to automatically learn what kind of activities, where and how, are performed by classes of entities from sparse argument co-occurrences with a verb; this information it is useful for sentence reconstruction tasks. Calvo et al. (2009b) propose a non language-dependent model based on the Word Space Model for calculating the plausibility of candidate argume...

Towards geospatial semantic search: exploiting latent semantic relations in geospatial data Wenwen Li a , Michael F. Goodchild b & Robert Raskin c a GeoDa Center for Geospatial Analysis and Computation, School of Geographical Science and Urban Planning, Arizona State University, Tempe, AZ, USA b Center for Spatial Studies (Spatial@UCSB), University of California, Santa Barbara, CA, USA c NASA J...

Literature search as a fundamental, complex and time-consuming step in a literature research process is part of many established scientific methods. It is still predominantly supported by search techniques based on conventional term-matching methods. We address the lack of semantic approaches in this context by proposing an enhancement of the literature research process with a prototype of our ...

A new method for automatic indexing and retrieval is described. The approach is to take advantage of implicit higher-order structure in the association of terms with documents ("semantic structure") in order to improve the detection of relevant documents on the basis of terms found in queries. The particular technique used is singular-value decomposition, in which a large term by document matri...

This paper proposes and examines modifications for the method of Latent Semantic Analysis (LSA). Several new local and global weight functions, along with normalization routines, are disclosed. Changes in the general structure of LSA are discussed. An application of LSA, in which the method is used to filter advertisements in e-mail, proves the worthiness of the advancements.

In latent semantic analysis (LSA), we aim at modelling a large corpus of high-dimensional discrete data from probabilistic perspective. The Assumption: one data point can be modelled by latent factors, which account for the co-occurrence of items within the data. We are also interested in the clustering structure of the data, which may benefit from the latent factors of the items. For example: ...