Revue: | Computación y sistemas |
Base de datos: | |
Número de sistema: | 000560508 |
ISSN: | 1405-5546 |
Autores: | Ramos, J. Guadalupe2 Ferreira Escutia, Rogelio1 Ferreira Medina, Heberto3 |
Instituciones: | 1Instituto Tecnológico de Morelia, Instituto Tecnológico de Morelia, Morelia, Michoacán. México 2Université Paris 13, París. Francia 3Universidad Nacional Autónoma de México, Instituto de Investigaciones en Ecosistemas y Sustentabilidad, Morelia, Michoacán. México |
Año: | 2020 |
Periodo: | Abr-Jun |
Volumen: | 24 |
Número: | 2 |
Paginación: | 703-714 |
País: | México |
Idioma: | Inglés |
Tipo de documento: | Artículo |
Resumen en inglés | For many web sites, money earning is crucial for keeping the content production. Advertising is one of the most common strategies for web content monetization. To determine where is the more appropriate place for ad location is an essential task to get a gentle introduction of the commercial information. There are algorithms based on keywords appearing within text; however we consider that implicit meaning is more adequate for a better harmony between content advertising. In this work, we present a formal method that determines the best place for advertising location. For this, we explore the underlying tree-like structure of a web page, we extract the text from each (X)HTML node and compute the semantic similarity (by employing latent semantic analysis) w.r.t., the advertising source text. We introduce a unique formula for the numerical calculation of the web page node relevance. We think it could be used for measuring the concordance among web page nodes and the commercial information and for the design of dynamic ads insertion methods. |
Disciplinas: | Ciencias de la computación |
Palabras clave: | Procesamiento de datos |
Keyword: | Semantic similarity, Latent semantic analysis, Dynamic advertising, Data processing |
Texte intégral: | Texto completo (Ver HTML) Texto completo (Ver PDF) |