Evaluation of Hybrid Collaborative Filtering Approach with Context-Sensitive Recommendation System

Murad, Dina Fitria; Hassan, Rosilah; Wijanarko, Bambang; Leandros, Riyan; Murad, Silvia Ayunda

doi:10.1109/icbir54589.2022.9786506

Cited by 10 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, one disadvantage of CF algorithms is that they may produce low-quality suggestions when dealing with sparse data, which is also referred to as the "cold-start problem", as stated above. To address this problem, a hybrid method [17] is advised, which entails including certain elements of content-based filtering into the recommendation development process for the initial user profile.…”

Section: Introductionmentioning

confidence: 99%

Harnessing the Power of User-Centric Artificial Intelligence: Customized Recommendations and Personalization in Hybrid Recommender Systems

et al. 2023

View full text Add to dashboard Cite

Recommender systems are widely used in various fields, such as e-commerce, entertainment, and education, to provide personalized recommendations to users based on their preferences and/or behavior. Τhis paper presents a novel approach to providing customized recommendations with the use of user-centric artificial intelligence. In greater detail, we introduce an enhanced collaborative filtering (CF) approach in order to develop hybrid recommender systems that personalize search results for users. The proposed CF enhancement incorporates user actions beyond explicit ratings to collect data and alleviate the issue of sparse data, resulting in high-quality recommendations. As a testbed for our research, a web-based digital library, incorporating the proposed algorithm, has been developed. Examples of operation of the use of the system are presented using cognitive walkthrough inspection, which demonstrates the effectiveness of the approach in producing personalized recommendations and improving user experience. Thus, the hybrid recommender system, which is incorporated in the digital library, has been evaluated, yielding promising results.

show abstract

Section: Introductionmentioning

confidence: 99%