Extending collaborative filtering recommendation using word embedding: A hybrid approach

2021 ◽  
Author(s):  
Luong Vuong Nguyen ◽  
Tri‐Hai Nguyen ◽  
Jason J. Jung ◽  
David Camacho
Keyword(s):  
Collaborative Filtering ◽  
Hybrid Approach ◽  
Word Embedding

A hybrid approach for article recommendation in research social networks

2017 ◽  
Vol 44 (5) ◽  
pp. 696-711 ◽  
Author(s):  
Jianshan Sun ◽  
Yuanchun Jiang ◽  
Xusen Cheng ◽  
Wei Du ◽  
Yezheng Liu ◽  
...  
Keyword(s):  
Social Networks ◽  
Social Network ◽  
Social Networking ◽  
Collaborative Filtering ◽  
Digital Libraries ◽  
User Study ◽  
Hybrid Approach ◽  
Quality Information ◽  
Complex Task ◽  
Data Sparsity

With the prevalence of research social networks, determining effective methods for recommending scientific articles to online scholars has become a challenging and complex task. Current studies on article recommendation works are focused on digital libraries and reference sharing websites while studies on research social networking websites have seldom been conducted. Existing content-based approaches or collaborative filtering approaches suffer from the problem of data sparsity. The quality information of articles has been largely ignored in previous studies, thus raising the need for a unified recommendation framework. We propose a hybrid approach to combine relevance, connectivity and quality to recommend scientific articles. The effectiveness of the proposed framework and methods is verified using a user study on a real research social network website. The results demonstrate that our proposed methods outperform baseline methods.

scholarly journals A Hybrid Ontology-Based Recommendation System in e-Commerce

Algorithms ◽  
2019 ◽  
Vol 12 (11) ◽  
pp. 239 ◽  
Author(s):  
Márcio Guia ◽  
Rodrigo Rocha Silva ◽  
Jorge Bernardino
Keyword(s):  
Collaborative Filtering ◽  
Experimental Evaluation ◽  
Recommendation System ◽  
Knowledge Engineering ◽  
Hybrid Approach ◽  
Recommendation Systems ◽  
The Internet ◽  
Active User ◽  
Expensive Process

The growth of the Internet has increased the amount of data and information available to any person at any time. Recommendation Systems help users find the items that meet their preferences, among the large number of items available. Techniques such as collaborative filtering and content-based recommenders have played an important role in the implementation of recommendation systems. In the last few years, other techniques, such as, ontology-based recommenders, have gained significance when reffering better active user recommendations; however, building an ontology-based recommender is an expensive process, which requires considerable skills in Knowledge Engineering. This paper presents a new hybrid approach that combines the simplicity of collaborative filtering with the efficiency of the ontology-based recommenders. The experimental evaluation demonstrates that the proposed approach presents higher quality recommendations when compared to collaborative filtering. The main improvement is verified on the results regarding the products, which, in spite of belonging to unknown categories to the users, still match their preferences and become recommended.

scholarly journals A Personalized Recommendation Algorithm Based on the User’s Implicit Feedback in E-Commerce

2018 ◽  
Vol 10 (12) ◽  
pp. 117 ◽  
Author(s):  
Bo Wang ◽  
Feiyue Ye ◽  
Jialu Xu
Keyword(s):  
Collaborative Filtering ◽  
Recommendation System ◽  
Word Embedding ◽  
Purchase Behavior ◽  
Personalized Recommendation ◽  
Implicit Feedback ◽  
Recommendation Algorithm ◽  
Rating Data ◽  
Feedback Information ◽  
Recommendation Accuracy

A recommendation system can recommend items of interest to users. However, due to the scarcity of user rating data and the similarity of single ratings, the accuracy of traditional collaborative filtering algorithms (CF) is limited. Compared with user rating data, the user’s behavior log is easier to obtain and contains a large amount of implicit feedback information, such as the purchase behavior, comparison behavior, and sequences of items (item-sequences). In this paper, we proposed a personalized recommendation algorithm based on a user’s implicit feedback (BUIF). BUIF considers not only the user’s purchase behavior but also the user’s comparison behavior and item-sequences. We extracted the purchase behavior, comparison behavior, and item-sequences from the user’s behavior log; calculated the user’s similarity by purchase behavior and comparison behavior; and extended word-embedding to item-embedding to obtain the item’s similarity. Based on the above method, we built a secondary reordering model to generate the recommendation results for users. The results of the experiment on the JData dataset show that our algorithm shows better improvement in regard to recommendation accuracy over other CF algorithms.

Personalisation of Advertisements in the Digital TV Context

2011 ◽  
pp. 264-284 ◽  
Author(s):  
George D. Lekakos ◽  
George M. Giaglis
Keyword(s):  
Collaborative Filtering ◽  
Hybrid Approach ◽  
Sparse Data ◽  
Digital Tv ◽  
Prediction Methods ◽  
Inherent Limitation ◽  
Rapid Changes ◽  
Filtering Technique ◽  
Stable Characteristic ◽  
Unique Domain

In this chapter, we discuss personalisation of advertisements in the digital TV environment and propose an effective personalisation approach, taking into account unique domain requirements. The proposed approach combines the widely used Pearson-based collaborative filtering technique, applied on numerical ratings with the user’s lifestyle, a stable characteristic drawn from consumer behaviour theory. We claim that users with similar lifestyles are reliable neighbours and can be utilised for the recommendation of advertisements for any member of their lifestyle neighbourhood. We focus on an inherent limitation of collaborative filtering methods that occurs when few ratings are available for each user and demonstrate that the proposed approach effectively manages this problem. Indeed, the hybrid approach combines the ability of the Pearson-based approach to accommodate rapid changes in user needs and make predictions upon one-click interactions and the advantage of the lifestyle-based approach to handle sparse data, which significantly affects the performance of collaborative filtering prediction methods.

Sign in / Sign up

Export Citation Format

Share Document