Content-based tag propagation and tensor factorization for personalized item recommendation based on social tagging

Rafailidis, Dimitrios; Αξενόπουλος, Απόστολος; Etzold, Jonas; Manolopoulou, Stavroula; Daras, Petros

doi:10.1145/2487164

Cited by 22 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wikipedia and Delicious are widely used by researchers to evaluate tag recommendation systems (Rafailidis et al , 2014; Sordo et al , 2013; Subramaniyaswamy and Chenthur Pandian, 2012). To measure the efficiency of MFS-LDA we used a publicly available data set (Zubiaga, 2012) of 20,578 Wikipedia articles annotated by the Delicious community.…”

Section: Experiments and Resultsmentioning

confidence: 99%

MFS-LDA: a multi-feature space tag recommendation model for cold start problem

Masood

Abbasi

Maqbool

et al. 2017

PROG

View full text Add to dashboard Cite

Purpose Tags are used to annotate resources on social media platforms. Most tag recommendation methods use popular tags, but in the case of new resources that are as yet untagged (the cold start problem), popularity-based tag recommendation methods fail to work. The purpose of this paper is to propose a novel model for tag recommendation called multi-feature space latent Dirichlet allocation (MFS-LDA) for cold start problem. Design/methodology/approach MFS-LDA is a novel latent Dirichlet allocation (LDA)-based model which exploits multiple feature spaces (title, contents, and tags) for recommending tags. Exploiting multiple feature spaces allows MFS-LDA to recommend tags even if data from a feature space is missing (the cold start problem). Findings Evaluation of a publicly available data set consisting of around 20,000 Wikipedia articles that are tagged on a social bookmarking website shows a significant improvement over existing LDA-based tag recommendation methods. Originality/value The originality of MFS-LDA lies in segregation of features for removing bias toward dominant features and in synchronization of multiple feature space for tag recommendation.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

MFS-LDA: a multi-feature space tag recommendation model for cold start problem

Masood

Abbasi

Maqbool

et al. 2017

PROG

View full text Add to dashboard Cite

show abstract

“…In the future, to test this idea we plan to extend our SDPL model to generate recommendations for Netflix users based on user data from Epinions and Slashdot [46]- [49]. In addition, we plan to explore ways to extend the proposed model to account for evolving user preferences [50]- [53], hybrid recommendations [54] and social event detection [55].…”

Section: Discussionmentioning

confidence: 99%

Learning to Rank with Trust and Distrust in Recommender Systems

Rafailidis

Crestani

2017

Proceedings of the Eleventh ACM Conference on Recommender Systems

Self Cite

View full text Add to dashboard Cite

The data scarcity of user preferences and the coldstart problem often appear in real-world applications and limit the recommendation accuracy of collaborative filtering strategies. Leveraging the selections of social friends and foes can efficiently face both problems. In this study, we propose a strategy that performs social deep pairwise learning. Firstly, we design a ranking loss function incorporating multiple ranking criteria based on the choice in users, and the choice in their friends and foes to improve the accuracy in the top-k recommendation task. We capture the nonlinear correlations between user preferences and the social information of trust and distrust relationships via a deep learning strategy. In each backpropagation step, we follow a social negative sampling strategy to meet the multiple ranking criteria of our ranking loss function. We conduct comprehensive experiments on a benchmark dataset from Epinions, among the largest publicly available that has been reported in the relevant literature. The experimental results demonstrate that the proposed model beats other state-of-the art methods, attaining an 11.49% average improvement over the most competitive model. We show that our deep learning strategy plays an important role in capturing the nonlinear correlations between user preferences and the social information of trust and distrust relationships, and demonstrate the importance of our social negative sampling strategy on the proposed model.

show abstract

“…Liu et al described a method that first divides unlabelled data points into clusters, selects exemplars from each cluster for the user to label, and propagates the labels to other data points in the cluster [23]. Rafailidis et al described the content-based tag propagation technique that propagates user-provided tags to similar items to address the "cold start" problem and boost the accuracy of tag-based search engine [36]. Kucher described ALVA, an interactive classification technique for text data annotation and visualization of the annotation.…”

Section: Interactive Classification With Intelligent User Interfacesmentioning

confidence: 99%

MI3: Machine-initiated Intelligent Interaction for Interactive Classification and Data Reconstruction

Zhang

CoeckeBob

ChenMin

2021

ACM Trans. Interact. Intell. Syst.

View full text Add to dashboard Cite

In many applications, while machine learning (ML) can be used to derive algorithmic models to aid decision processes, it is often difficult to learn a precise model when the number of similar data points is limited. One example of such applications is data reconstruction from historical visualizations, many of which encode precious data, but their numerical records are lost. On the one hand, there is not enough similar data for training an ML model. On the other hand, manual reconstruction of the data is both tedious and arduous. Hence, a desirable approach is to train an ML model dynamically using interactive classification, and hopefully, after some training, the model can complete the data reconstruction tasks with less human interference. For this approach to be effective, the number of annotated data objects used for training the ML model should be as small as possible, while the number of data objects to be reconstructed automatically should be as large as possible. In this article, we present a novel technique for the machine to initiate intelligent interactions to reduce the user’s interaction cost in interactive classification tasks. The technique of machine-initiated intelligent interaction (MI3) builds on a generic framework featuring active sampling and default labeling. To demonstrate the MI3 approach, we use the well-known cholera map visualization by John Snow as an example, as it features three instances of MI3 pipelines. The experiment has confirmed the merits of the MI3 approach.

show abstract

Content-based tag propagation and tensor factorization for personalized item recommendation based on social tagging

Cited by 22 publications

References 36 publications

MFS-LDA: a multi-feature space tag recommendation model for cold start problem

MFS-LDA: a multi-feature space tag recommendation model for cold start problem

Learning to Rank with Trust and Distrust in Recommender Systems

MI3: Machine-initiated Intelligent Interaction for Interactive Classification and Data Reconstruction

Contact Info

Product

Resources

About