A Web Mining Approach for Personalized E-Learning System

Abstract

Abstract-The Web Mining plays a very important role for the E-learning systems. In personalized E-Learning system, user customize the learning environment based on personal choices. In a general search process ,a hyperlink which is having maximum number of hits will get displayed first . For making a personalized system history of every user need to be saved in the form of user logs. In this paper we present a architecture with the use of Web mining for Web personalization. The proposed system provides a new approach with combination of web usage mining, HITS algorithm and web content mining. It combines hits results on user logs and web page contents with a clustering algorithm called as Lingo clustering algorithm. This proposed system with combined approach gives a better performance than a usage based system. Further the results are computed according to matrices computed from previous and proposed method.

User’s Learning Capability Aware E-Content Recommendation System For Enhanced Learning Experience

“…In [20], the server-side user logs were kept to analyze user interest. This gave browsing-based recommendations.…”

“…Precision: Precision is determined by comparing the number of documents that fall under the class i but were incorrectly classified by the proposed work with label j to the total number of documents that have the label j. This is computed by (20).…”

“…𝑑𝑜𝑐 𝑖𝑗 𝑑𝑜𝑐 𝑗 (20) where 𝑑𝑜𝑐 𝑖𝑗 is the total number of documents that should be categorized as belonging to class 𝑖, but were instead assigned to class 𝑗, due to an error in the classification process. 𝑑𝑜𝑐 𝑗 is the number of documents that may be categorized as belonging to the label 𝑗.…”

Modelación matemática para optimizar la estructura de los cursos en las plataformas e-learning

“…La nueva estructura propuesta permite la optimización de los cursos de las plataformas elearning a través de la información almacenada en los ficheros log, lo cual complementa los estudios realizados por los autores: (Zaíane and Luo, 2001;Romero et al, 2009a;Lorena, 2010;Aponte et al, 2012;Ballesteros et al, 2013 y Chakurkar andAdiga, 2015).…”