Point-of-Interests Recommendation Service in Location-Based Social Networks: A Survey, Research Challenges, and Future Perspectives

Asaad, Safar Maghdid; Ghafoor, Kayhan Zrar; Sarhang, Halgurd; Mulahuwaish, Aos

doi:10.1007/978-3-031-08815-5_4

Cited by 1 publication

(1 citation statement)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiple recommendation approaches can be employed to conceal the flaws of each individual recommendation strategy in a combined model. Integrating the techniques may be carried out in a variety of ways, including individually applying the techniques and combining the results [22]. There are seven basic hybridization techniques of different recommendation method combinations: weighted, switching, mixed, feature combination, feature augmentation, cascade, and meta-level.…”

Section: ) Collaborative Filtering Techniquementioning

confidence: 99%

An e-Learning Recommendation System Framework

Kong,

Haw,

Palanichamy

et al. 2024

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

With the emergence of the digital era, the e-learning platform has become an effective tool for obtaining quality e-learning content. However, despite its potential, the true extent of its capabilities has yet to be fully explored. In order to attract users and maximize revenue, e-learning platforms are now expected to provide content tailored to their users' needs and preferences. These recommendations are generated by considering factors such as prior purchases, browsing history, demographic information, and more. By leveraging these advanced technologies, e-learning platforms can enhance the learning experience by providing users with content that is both engaging and relevant to their individual needs and interests. This paper explores the popular Machine Learning (ML) techniques employed in e-learning content recommender platforms. Two machine learning techniques, k-Nearest Neighbour Baseline (KNNBaseline) and Singular Value Decomposition (SVD), are selected and used to accurately forecast customer interests and preferences. By examining the data patterns and user behaviors, these ML techniques provide insights into the most relevant and personalized educational content for individual users, enhancing their learning experience. The item ratings predicted are generated based on the underlying pattern in past ratings of users. The performance of applied approaches was assessed using several evaluation metrics, which include root mean square error and mean absolute error.

show abstract

Section: ) Collaborative Filtering Techniquementioning

confidence: 99%