RETRACTED ARTICLE: Customer centric hybrid recommendation system for E-Commerce applications by integrating hybrid sentiment analysis

Arodh Lal Karn (), Rakshha Kumari Karna (), Bhavana Raj Kondamudi (), Girish Bagale (), Denis A. Pustokhin (), Irina V. Pustokhina () and Sudhakar Sengan ()
Additional contact information
Arodh Lal Karn: Xi’an Jiaotong-Liverpool University
Rakshha Kumari Karna: Harbin Institute of Technology
Bhavana Raj Kondamudi: Institute of Public Enterprise
Girish Bagale: SVKM’s NMIMS University
Denis A. Pustokhin: State University of Management
Irina V. Pustokhina: Plekhanov Russian University of Economics
Sudhakar Sengan: PSN College of Engineering and Technology

Electronic Commerce Research, 2023, vol. 23, issue 1, No 13, 279-314

Abstract: Abstract Internet applications such as Online Social Networking and Electronic commerce are becoming incredibly common, resulting in more content being available. Recommender systems (RS) assist users in identifying appropriate information out of a large pool of options. In today’s internet applications, RS are extremely important. To increase user satisfaction, this type of system supports personalized recommendations based on a massive volume of data. These suggestions assist clients in selecting products, while concerns can increase product utilization. We discovered that much research work is going on in the field of recommendation and that there are some effective systems out there. In the context of social information, sentimental analysis (SA) can aid in increasing the knowledge of a user’s behaviour, views, and reactions, which is helpful for incorporating into RS to improve recommendation accuracy. RS has been found to resolve information overload challenges in information retrieval, but they still have issues with cold-start and data sparsity. SA, on the other hand, is well-known for interpreting text and conveying user choices. It’s frequently used to assist E-Commerce in tracking customer feedback on their products and attempting to comprehend customer needs and preferences. To improve the accuracy and correctness of any RS, this paper proposes a recommendation model based on a Hybrid Recommendation Model (HRM) and hybrid SA. In the proposed method, we first generate a preliminary recommendation list using a HRM. To generate the final recommendation list, the HRM with SA is used. In terms of various evaluation criteria, the HRM with SA outperforms traditional models.

Keywords: Hybrid recommender system; E-Commerce; Sentiment analysis; Neural network; Accuracy; Filtering; Deep learning; Bert (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://link.springer.com/10.1007/s10660-022-09630-z Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:elcore:v:23:y:2023:i:1:d:10.1007_s10660-022-09630-z

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10660

DOI: 10.1007/s10660-022-09630-z

Access Statistics for this article

Electronic Commerce Research is currently edited by James Westland

More articles in Electronic Commerce Research from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().