Enhancing the accuracy of group recommendation using slope one

• Published in: The Journal of supercomputing 2023, Vol.79 (1), p.499-540
• Main Authors: Yannam, V Ramanjaneyulu, Kumar, Jitendra, Babu, Korra Sathya, Patra, Bidyut Kumar
• Format: Article
• Language: English
• Subjects: Compilers; Computer Science; Interpreters; Performance prediction; Processor Architectures; Programming Languages; Recommender systems
• ISSN: 0920-8542; 1573-0484
• DOI: 10.1007/s11227-022-04664-4

Recommender systems provide personalized suggestions to users regarding products and services. These recommendations are generated for individual users only. However, group activities are gaining popularity in several applications such as movie recommendations, e-tourist, etc. There is a need for a group recommendation system that helps people to provide better recommendations for group members instead of a single user. Most of the works in group recommendation systems use two approaches as group aggregate prediction and group aggregate model. Group aggregate prediction performs based on the group individual prediction information. The group aggregate model is determined by aggregating individual user preferences. Accurate prediction for a group using the choices of other group members is a challenging task, and this prediction task was performed using a static approach. In this work, we propose group recommendations using Slope One for generating group recommendations. Slope One is capable of producing relevant results in a shorter time and is efficient at query time; one of the advantages is to support the dynamic updates of the rating predictions in online mode. Also, a novel modelling technique (Max after threshold) is introduced. It adopts an aggregate prediction feature to provide better recommendations to the group by using Slope One. The experiments are done on two benchmark datasets. The proposed technique outperforms the state-of-the-art group recommendation models.