Vulcanus 2.0: A Recommender System for Accessibility

CLEI Electronic Journal

Jorge Luis Victoria Barbosa¹  Rodrigo da Rosa Righi¹  Bruno Mota²  Ismael Gomes Cardoso² 

[1] Applied Computing Graduate Program - Unisinos Av. Unisinos, 950 – São Leopoldo, RS, Brazil;University of Vale do Rio dos Sinos, Applied Computing Graduate Program, Av. Unisinos, 950 - São Leopoldo, RS, Brazil;

关键词: Recommender System;    Accessibility;    Context Awareness;    Asymptotic Analysis;   

DOI  :  10.19153/cleiej.19.1.6

来源: DOAJ

【 摘 要 】

Even though the use of recommender systems is already widely spread in several application areas, there is still a lack of studies for accessibility research field. One of these attempts to use recommender system benefits for accessibility needs is Vulcanus. The Vulcanus recommender system uses similarity analysis to compare user’s trails. In this way, it is possible to take advantage of the user’s past behavior and distribute personalized content and services. The Vulcanus combined concepts from ubiquitous computing, such as user profiles, context awareness, trails management, and similarity analysis. It uses two different approaches for trails similarity analysis: resources patterns and categories patterns. In this work we performed an asymptotic analysis, identifying Vulcanus’ algorithm complexity. Furthermore we also propose improvements achieved by dynamic programming technique, so the ordinary case is improved by using a bottom-up approach. With that approach, many unnecessary comparisons can be skipped and now Vulcanus 2.0 is presented with improvements in its average case scenario.

【 授权许可】

Unknown