SVD-based collaborative filtering with privacy

Polat, Hüseyin; Du, Wenliang

doi:10.1145/1066677.1066860

Cited by 172 publications

(106 citation statements)

References 13 publications

Supporting

Mentioning

106

Contrasting

Order By: Relevance

“…Matrix decomposition method is a commonly used matrix dimension reduction method, and it also has the function of data noise filtering. Singular Value Decomposition(SVD) is a commonly used matrix decomposition technique which can effectively extract matrix eigenvalue, reveal the internal structure of the matrix, and also avoid excessive sparse data in a certain degree [11] . SVD divides a m×n matrix R into 3 matrices.…”

Section: Svd (Single Value Decomposition)mentioning

confidence: 99%

See 1 more Smart Citation

A Hybrid Recommendation Algorithm Based On SVD And Weighted Slope One

Zhou¹,

Yu²

2015

Proceedings of the International Conference on Management, Computer and Education Informatization

View full text Add to dashboard Cite

Abstract-With the arrival of the era of big data, recommendation algorithm has become a hot research field in internet life and social business. However, the traditional recommendation algorithms have problems of cold start, scoring matrix data sparse, affecting the accuracy of recommendation. This paper proposes a hybrid recommendation algorithm based on the SVD (singular value decomposition) and Slope One algorithm, experiments results show that the algorithm proposed by this paper has superior performance compared with the traditional collaborative filtering recommendation algorithm.

show abstract

Section: Svd (Single Value Decomposition)mentioning

confidence: 99%

“…It judges the accuracy of algorithm by calculating the deviation between the predictive interest of the user to an item and the real interest [9]. Set the actual score set for {q1,q2,…,qN}, predictive score set for{p1,p2,…,pN}, then the MAE can be defined as (11) …”

Section: ) Mae(mean Absolute Error)mentioning

confidence: 99%

A Hybrid Recommendation Algorithm Based On SVD And Weighted Slope One

Zhou¹,

Yu²

2015

Proceedings of the International Conference on Management, Computer and Education Informatization

View full text Add to dashboard Cite

show abstract

“…Randomized Response technique was first introduced by Warner as a technique to solve the following survey problem: To estimate the percentage of people in a population that has attribute A, ueries are sent to a group of people (Polat and Du, 2005). Since the attribute A is related to some confidential aspects of human life, respondents may decide not to reply at all or to reply with incorrect answers.…”

Section: Randomized Response Techniquesmentioning

confidence: 99%

Secured Disclosure of Sensitive Data in Data Mining Techniques

Gurusamy¹

2012

Journal of Computer Science

View full text Add to dashboard Cite

order to produce meaningful results. Based on the analysis of shortcomings of earlier technologies this study proposes a new method for securing numerical and categorical data. In this method the categorical data is converted into Binary form and perturbation based noise is introduced as a security method based on the security level anticipated. Several types of noise addition methods were employed and generalized results were evaluated in terms of misclassification error and privacy level. An average of misclassification error was below 50% for 75-90% security level, which is better than earlier methods which didn't handle categorical data. The results obtained prove that the proposed method outperforms some of the currently existing methods thereby ensuring the possibility of securing sensitive data irrespective of its type being numerical or categorical.

show abstract

“…Polat and Du (Polat & Du, 2003, 2005a, 2005b demonstrate the usage of randomized perturbation techniques (adding random numbers from a given range to the original data) in disguising the original user ratings before feeding them into CF algorithms based on correlation and singular value decomposition. The CF system thereby does not know the exact values of the original ratings, yet is still able to compute reasonably accurate recommendations.…”

Section: Randomized Perturbationmentioning

confidence: 99%

Technical Solutions for Privacy- Enhanced Personalization

Wang

2009

Intelligent User Interfaces

View full text Add to dashboard Cite

This chapter presents a first-of-its-kind survey that systematically analyzes existing privacy-enhanced personalization (PEP) solutions and their underlying privacy protection techniques. The evaluation is based on an analytical framework of privacy-enhancing technologies, an earlier work of the authors. More specifically, we critically examine whether each PEP solution satisfies the privacy principles and addresses the privacy concerns that have been uncovered in the context of personalization. The chapter aims at helping researchers better understand the technical underpinnings, practical efficacies and limitations of existing PEP solutions, and at inspiring and developing future PEP solutions by outlining several promising research directions based on our findings.

show abstract

SVD-based collaborative filtering with privacy

Cited by 172 publications

References 13 publications

A Hybrid Recommendation Algorithm Based On SVD And Weighted Slope One

A Hybrid Recommendation Algorithm Based On SVD And Weighted Slope One

Secured Disclosure of Sensitive Data in Data Mining Techniques

Technical Solutions for Privacy- Enhanced Personalization

Contact Info

Product

Resources

About