Improvement of collaborative filtering using rating normalization

Kim, Soo-Cheol; Sung, Kyoung-Jun; Park, Chansoo; Kim, Sung Kwon

doi:10.1007/s11042-013-1814-0

Cited by 22 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normalization is to map the data of each behavior to a specified range and then process it, so that the overall preference obtained by weighted summation would become more accurate. Normalization can process data more conveniently and quickly, moreover, it also enables objects of different units or magnitudes to be compared and weighted [6]. Normalization not only simplifies the complex calculation, but also furtherly excavates the available data.…”

Section: Process Of Collaborative Filterinngmentioning

confidence: 99%

Comparison Between Collaborative Filtering and Content-Based Filtering

Guo¹

2022

HSET

View full text Add to dashboard Cite

With the rapid development of Internet technology nowadays, how to quickly obtain the effective information needed by users has become the key point of the scientific and technological academia. Therefore, various kinds of recommendation algorithms have been invented. Based on the previous research, this paper introduces the most famous and widely used recommendation algorithms among many recommendation systems, which are collaborative filtering and content-based filtering. In this paper, the core ideas and operation principles of the two algorithms are introduced in detail. In addition, by describing the steps of these two algorithms gradually and analyzing their processes step by step, we can accurately analyze and summarize their advantages and disadvantages respectively. And on this basis, the respective areas which they are good at are mentioned. Moreover, this paper points out the shortcomings and limitations that still exist at present, and the direction for further improvement in the future. Finally, at the end of the paper, there are some overall comparation and summation about the two algorithms. And the hot research points of them in the future are discussed.

show abstract

Section: Process Of Collaborative Filterinngmentioning

confidence: 99%

Comparison Between Collaborative Filtering and Content-Based Filtering

Guo¹

2022

HSET

View full text Add to dashboard Cite

show abstract

“…User behaviour was not independent; it followed the trend of others. Kim et al (2016) proposed an improvement of an existing preference prediction algorithm to increase the accuracy of recommendation systems. In a recommendation system, prediction of items preferred by users was based on their ratings.…”

Section: Related Workmentioning

confidence: 99%

Collaborative filtering algorithm based on multi-factors

Wang

Yuan

2020

IJCSM

View full text Add to dashboard Cite

Recommender systems are widely used to provide e-commerce users appropriate items and have emerged in response to the problem of information overload. Collaborative filtering (CF) is one of the most successful recommender methods which recommend items to a given user based on the opinions of the similar users. However, the existing CF methods lack the consideration of factors such as time and geo-location. In this paper, we take into account many influencing factors including time and geo-location in the process of similarity computation. The simulation results on two real-world data sets show that our algorithm achieves superior performance to existing methods.

show abstract

“…The second motivation is to overcome the limitations and drawbacks suffered by most of the existing recommender systems. In which the algorithms adopted only rely on the extremely sparse user-item rating data, and the computed similarity in collaborative filtering based recommender systems will be potentially unreliable and incorrect [16,20]. For example, new users can not get their favorite items until the systems have acquired the preference from their historical behavior records [21], and new items cannot be recommended to the suitable users until enough users have behaviors (browse, download, give a mark) on them.…”

Section: Bipartite Graphical Correlation and Implicit Trust Based Recmentioning

confidence: 99%

“…It can also be found from the results that although the bipartite correlation and implicit trust both have influences on acquiring the preferences, the bipartite correlation reflect users' preference more accurately. We compare our algorithm with RNCF [16] and GBR [19] in the test data set of MovieLens and conceal the users' data randomly by ABO. The GBR uses directed trust graph to generate recommendation which is one of the latest applications of graph model on recommender systems.…”

Section: Experiments 1 Impact Of Independent Trust Groupmentioning

confidence: 99%

“…Zhao et al [15] proposed a ranking algorithmic framework for heterogeneous manifold in social tagging services. Kim et al [16] used rating normalization to improve collaborative filtering based recommender systems. In these methods mentioned above, users' historical preference is often described as user-item rating matrix, and the prediction is generated by process and analysis of the matrix.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A preference elicitation method based on bipartite graphical correlation and implicit trust

Gao

Fan

et al. 2017

Neurocomputing

View full text Add to dashboard Cite

In the age of big data, information overload is getting worse. Most of the existing recommender systems which apply data analysis and behavioral analysis to make personal recommendation have suffered from the problem of low prediction accuracy. To address this problem, a new preference elicitation algorithm is developed by using bipartite graphical correlation and implicit trust. More precisely, to compute the bipartite graphical correlation, an improved single-source shortest path is firstly presented to get the shortest behavior path based on the user-item bipartite graph. While the Markov separation algorithm is employed to separate uncorrelated vertexes and obtain the values of bipartite graphical correlation. Secondly, the implicit trust is used to represent trust relationship between users and items which have no historical behaviors. Independent trust group is applied to represent the closed-loop path on the user-item bipartite graph, and a compute method is developed to get the values of implicit trust based on the trusty values of independent trust group and explicit trust. Finally, a weighted prediction method based on bipartite graphical correlation and implicit trust is employed to compute the final ratings. Our empirical experiments are performed on a sparse data set, the results of which have demonstrated that our method can achieve lower P@R and efficiently improve recommendation quality.

show abstract

Improvement of collaborative filtering using rating normalization

Cited by 22 publications

References 16 publications

Comparison Between Collaborative Filtering and Content-Based Filtering

Comparison Between Collaborative Filtering and Content-Based Filtering

Collaborative filtering algorithm based on multi-factors

A preference elicitation method based on bipartite graphical correlation and implicit trust

Contact Info

Product

Resources

About