An impact of time and item influencer in collaborative filtering recommendations using graph-based model

Najafabadi, Maryam Khanian; Mohamed, Azlinah; Onn, Choo Wou

doi:10.1016/j.ipm.2018.12.007

Cited by 62 publications

(37 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our experiment, we used a pre-processing method that has been extensively used in many research works [12,13,[30][31][32]. In this method, the datasets are randomly divided into training and testing set, which were 80% of the ratings per user as training set and the remaining ratings (20%) as the test set.…”

Section: Experimental Methodologymentioning

confidence: 99%

“…Moreover, there are drawbacks of CF recommendation systems that need to be addressed in increasing the quality of recommendation and accuracy of the predicted rated. These drawbacks are high dimensionality, data sparsity, and cold-start [9][10][11][12]. Most of the proposed recommender systems in solving drawbacks of CF failed to take action based on both sides of similarity (similarity among users and items) and it was discovered that the amount of time spent in calculating similarity among users or items to produce recommendations was extended.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Effective Collaborative User Model Using Hybrid Clustering Recommendation Methods

Najafabadi¹,

Mohamed²,

Nair³

et al. 2021

ISI

Self Cite

View full text Add to dashboard Cite

Collaborative Filtering (CF) has been known as the most successful recommendation technique in which recommendations are made based on the past rating records from like-minded users. Significant growth of users and items have negatively affected the efficiency of CF and pose key issues related to computational aspects and the quality of recommendation such as high dimensionality and data sparsity. In this study, a hybrid method was proposed and was capable to solve the mentioned problems using a neighborhood selection process for each user through two clustering algorithms which were item-based k-means clustering and user-based Fuzzy Clustering. Item-based k-means clustering was applied because of its advantages in computational time and hence it is able to address the high dimensionality issues. To create user groups and find the correlation between users, we employed the user-based Fuzzy Clustering and it has not yet been used in user-based CF clustering. This clustering can calculate the degree of membership among users into set of clustered items. Furthermore, a new similarity metric was designed to compute the similarity value among users with affecting the output of user-based Fuzzy Clustering. This metric is an alternative to the basic similarity metrics in CF and it has been proven to provide high-quality recommendations and a noticeable improvement on the accuracy of recommendations to the users. The proposed method has been evaluated using two benchmark datasets, MovieLens and LastFM in order to make a comparison with the existing recommendation methods.

show abstract

Section: Experimental Methodologymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Effective Collaborative User Model Using Hybrid Clustering Recommendation Methods

Najafabadi¹,

Mohamed²,

Nair³

et al. 2021

ISI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Collaborative filtering is used to compare profiles of users in predicting possible interests of users based on similarity of preferences of ratings on items [ 22 ], while similarity measurement methods, such as the Pearson correlation coefficient, are also used to identify users based on ratings [ 23 ]. Furthermore, predicting users’ preferences can be identified using graph-based approaches and purchase time information [ 24 ].…”

Section: Related Workmentioning

confidence: 99%

Enhancing Personalized Ads Using Interest Category Classification of SNS Users Based on Deep Neural Networks

Hong

Choi

Lim

et al. 2020

Sensors

View full text Add to dashboard Cite

The classification and recommendation system for identifying social networking site (SNS) users’ interests plays a critical role in various industries, particularly advertising. Personalized advertisements help brands stand out from the clutter of online advertisements while enhancing relevance to consumers to generate favorable responses. Although most user interest classification studies have focused on textual data, the combined analysis of images and texts on user-generated posts can more precisely predict a consumer’s interests. Therefore, this research classifies SNS users’ interests by utilizing both texts and images. Consumers’ interests were defined using the Curlie directory, and various convolutional neural network (CNN)-based models and recurrent neural network (RNN)-based models were tested for our user interest classification system. In our hybrid neural network (NN) model, CNN-based classification models were used to classify images from users’ SNS postings while RNN-based classification models were used to classify textual data. The results of our extensive experiments show that the classification of users’ interests performed best when using texts and images together, at 96.55%, versus texts only, 41.38%, or images only, 93.1%. Our proposed system provides insights into personalized SNS advertising research and informs marketers on making (1) interest-based recommendations, (2) ranked-order recommendations, and (3) real-time recommendations.

show abstract

“…Therefore, Nzeko'O et al (Nzeko'O et al 2017) presented Time Weight Content-based STG to calculate the weightings of the edges based on their ages. Finally, Najafabadi et al (2019) proposed an algorithm that incorporates the visiting time into the graph structure and uses a hybrid CF algorithm to suggest top-N items to the users.…”

Section: Introductionmentioning

confidence: 99%

Point-of-interest recommendation using extended random walk with restart on geographical-temporal hybrid tripartite graph

Taheri

Farnaghi

Alimohammadi

et al. 2021

Journal of Spatial Science

View full text Add to dashboard Cite

Previous studies neglected the impact of shared spatio-temporal interests in the Point of Interest (POI) recommendation. The proposed GHTG-ERWR method simultaneously models shared spatio-temporal preferences of users, the dynamics of users' behavior, and the geographical effect for POI recommendation. It is composed of an Extended Random Walk with Restart (ERWR) algorithm that works on a Geographical-temporal Hybrid Tripartite Graph (GHTG). The method overcomes shortcomings of prior approaches by employing users' joint preferences over time and space through the auxiliary information derived from location-session and sessionlocation edges. Experimental results on Gowalla and Weeplaces datasets proved the feasibility of the approach.

show abstract

An impact of time and item influencer in collaborative filtering recommendations using graph-based model

Cited by 62 publications

References 26 publications

An Effective Collaborative User Model Using Hybrid Clustering Recommendation Methods

An Effective Collaborative User Model Using Hybrid Clustering Recommendation Methods

Enhancing Personalized Ads Using Interest Category Classification of SNS Users Based on Deep Neural Networks

Point-of-interest recommendation using extended random walk with restart on geographical-temporal hybrid tripartite graph

Contact Info

Product

Resources

About