New technique to alleviate the cold start problem in recommender systems using information from social media and random decision forests

Herce-Zelaya, Julio; Porcel, Carlos; Bernabé-Moreno, Juan; Tejeda-Lorente, Álvaro; Herrera‐Viedma, Enrique

doi:10.1016/j.ins.2020.05.071

Cited by 73 publications

(23 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where u u n denotes the number of ratings from the user u. According to Bayesian inference, the feature matrices U and V can be obtained by minimizing the following formula [39] .…”

Section: Recommendation Model Based On Neighborhood Relationshipsmentioning

confidence: 99%

A Comprehensive Social Matrix Factorization with Social Regularization for Recommendation Based on Implicit Similarity by Fusing Trust Relationships and Social Tags

Chen

Zhang

et al. 2021

Preprint

View full text Add to dashboard Cite

Social relationships play an important role in improving the quality of recommender systems (RSs). A large number of experimental results show that social relationship-based recommendation methods alleviate the problems of data sparseness and cold start in RSs to some extent. However, since the social relationships between users are extremely sparse and complex, and it is difficult to obtain accurately user preference model, thus the performance of the recommendation system is affected by the existing social recommendation methods. In order to accurately model social relationships and improve recommendation quality, we use explicit social relationships such as user-item ratings, trust relationships and implicit social relationships such as social tags to mine potential interest preferences of users and propose an improved social recommendation method integrating trust relationship and social tags. The method map user features and item features to the shared feature space by using the above social relationship, respectively, and obtains user similarity and item similarity through potential feature vectors of users and items, and continuously trains them to obtain accurate similarity relationship to improve the recommendation performance. Experimental results demonstrate that our proposed approach achieves superior performance to the other social recommendation approaches.

show abstract

“…where u u n denotes the number of ratings from the user u. According to Bayesian inference, the feature matrices U and V can be obtained by minimizing the following formula [39] .…”

Section: Recommendation Model Based On Neighborhood Relationshipsmentioning

confidence: 99%

A Comprehensive Social Matrix Factorization with Social Regularization for Recommendation Based on Implicit Similarity by Fusing Trust Relationships and Social Tags

Chen

Zhang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Similar to Ning's work, Dhelim et al [20] also uses Big-Five personality traits plus dynamic interest in its user interest mining system. Herce-Zelaya et al [21] presents profiles of user behavior using social media based on classification trees and random forests to create predictions and address cold start problems. Wang's research et al [22] introduced neighboring factors and time functions as well as utilizing dynamic selection models to select adjacent sets of objects.…”

Section: Research Related To Collaborative Filtering Recommendation Smentioning

confidence: 99%

Addressing Cold Start New User in Recommender System Based on Hybrid Approach: A review and bibliometric analysis

Ghifari

Sitohang

Saptawati

2021

ITJRD

View full text Add to dashboard Cite

Increasing number of internet users today, the use of e-commerce becomes a very vital need. One of the keys that holds the success of the e-commerce system is the recommendation system. Collaborative filtering is the popular method of recommendation system. However, collaborative filtering still has issues including data sparsity, cold start, gray sheep, and dynamic taste. Some studies try to solve the issue with hybrid methods that use a combination of several techniques. One of the studies tried to solve the problem by building 7 blocks of hybrid techniques with various approaches. However, the study still has some problems left. In the case of cold start new users, actually, the method in the study has handled it with matrix factorizer block and item weight. But it will produce the same results for all users so that the resulting personalization is still lacking. This study aims to map an overview of the themes of recommendation system research that utilizes bibliometric analysis to assess the performance of scientific articles while exposing solution opportunities to cold start problems in the recommendation system. The results of the analysis showed that cold start problems can be solved by utilizing social network data and graph approaches.

show abstract

“…Recommender systems use a class of information filter systems, that have the main goal to provide personalized recommendations, services, content to users [9]. The functionality of recommender systems focuses on information filter tools that aid users in their information access, through prediction and recommendations from history data patterns [10]. Based on the literature, there are many recommender system techniques applied in the industry.…”

Section: Recommender Systemmentioning

confidence: 99%

Interactive Recommender System for Identifying Learning Partner

Oktavia¹,

Sujarwo²

2021

IJETAE

View full text Add to dashboard Cite

— Currently, higher education must open their mindset to change the learning process. This process cannot still stay on a conventional process that only involved students and lecturers as learning participants because to capture what the current industry needs and the trend of knowledge, higher education institutions must collaborate with external parties as learning partners to give a global perspective about the industrial needs and trends. The process to identify a learning partner to contribute to the learning process is not easy way. The higher education institution must select which partner has appropriate skills and competency suitable for the subject’s course. Many parameters will be involved in the selection process to identify the right partner. In this proposed system, external parties in this context consist of professionals or educators from the external institution easily can be defined to be selected as learning partners based on their competency and experience, that listed on social media LinkedIn as a professional platform. The method to build this interactive recommender system in this study is based on design science research that identifies step by step design stage to propose the best way solution for the recommender system. The result of this study is an interactive recommender system that can help higher education to find out the best candidate for their learning partner so the collaboration of learning can be implemented effectively

show abstract

New technique to alleviate the cold start problem in recommender systems using information from social media and random decision forests

Cited by 73 publications

References 39 publications

A Comprehensive Social Matrix Factorization with Social Regularization for Recommendation Based on Implicit Similarity by Fusing Trust Relationships and Social Tags

A Comprehensive Social Matrix Factorization with Social Regularization for Recommendation Based on Implicit Similarity by Fusing Trust Relationships and Social Tags

Addressing Cold Start New User in Recommender System Based on Hybrid Approach: A review and bibliometric analysis

Interactive Recommender System for Identifying Learning Partner

Contact Info

Product

Resources

About