Estado de arte de "Latent Semantic Index"con una prueba experimental

This article describes Latent Semantic Index due to its specific purpose is unknown, this method has been used in several areas; the technique for their development uses an algorithm based on the handling of arrays called Singular Value Decomposition that has not been detailed in the literature...

Descripción completa

Autor Principal: Sosa Erazo, Marco Vinicio
Otros Autores: Zambonino Altamirano, Marilú Andrea
Formato: bachelorThesis
Idioma: spa
Publicado: 2018
Materias:
Acceso en línea: http://dspace.ups.edu.ec/handle/123456789/15283
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Sumario: This article describes Latent Semantic Index due to its specific purpose is unknown, this method has been used in several areas; the technique for their development uses an algorithm based on the handling of arrays called Singular Value Decomposition that has not been detailed in the literature review; based on the state of the art study the algorithm immersed in Latent Semantic Index for a documentary analysis on a corpus of documents; with the meaningful understanding of the topic, you can determine the how? and why? of Latent Semantix Index and its application in different areas; use the theoretical knowledge in the experimental test to take greater advantage of the method as a benefit in the management independent documentary of the field of application; the results are based on a process of vectors that compares the repetitions of each word in a document with the query to form a matrix; through the similarity of cosines of the vector space classic the documents are placed in descending order according to the percentage of relevance between the document and the search; the technique is mainly based on the quantification of each word in the document and not on the meaning of the word so the structure of the query does not affects the outcome.