A fuzzy model for managing natural noise in recommender systems

Toledo, Raciel Yera; Castro, Jorge; Martínez, Luis

doi:10.1016/j.asoc.2015.10.060

Cited by 51 publications

(39 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The best value for d 1 has been determined empirically [14], concluding that for d 1 = 1 the proposal obtains an optimal performance.…”

Section: Noise Detection Phasementioning

confidence: 99%

“…The management of NN depends on the data available and some approaches require additional user interactions [9] and others just remove noisy data either ratings or users [13]. Recently, however, new proposals manage the NN by using the current rating values in the user/item rating matrix while keeping as much information as possible and without any new user's interaction [10,14].…”

Section: Martin@ujaenesmentioning

confidence: 99%

“…Therefore, Yera, Castro and Martínez [14] focused on overcoming the lack of flexibility and robustness by proposing a novel fuzzy NN method to deal with the inherent uncertainty of NN, improving in such a way the noise management and consequently the recommendation accuracy. This approach extends the approach revised in Section 3 because it focuses on the NN management just using ratings.…”

Section: A Fuzzy Approach To Detect Noisy Ratingsmentioning

confidence: 99%

“…Several experiments [14] were performed to determine the most suitable value for the threshold, d 2 , determining that d 2 = 1 is the best value.…”

Section: Noise Detection Phasementioning

confidence: 99%

See 3 more Smart Citations

Managing Natural Noise in Recommender Systems

Martínez

Castro

Toledo

2016

Theory and Practice of Natural Computing

Self Cite

View full text Add to dashboard Cite

Abstract. E-commerce customers demand quick and easy access to suitable products in large purchase spaces. To support and facilitate this purchasing process to users, recommender systems (RSs) help them to find out the information that best fits their preferences and needs in an overloaded search space. These systems require the elicitation of customers' preferences. However, this elicitation process is not always precise either correct because of external factors such as human errors, uncertainty, human beings inherent inconsistency and so on. Such a problem in RSs is known as natural noise (NN) and can negatively bias recommendations, which leads to poor user's experience. Different proposals have been presented to deal with natural noise in RSs. Several of them require additional interaction with customers. Others just remove noisy information. Recently, new NN approaches dealing with the ratings stored in the user/item rating matrix have raised to deal with NN in a better and simpler way. This contribution is devoted to provide a brief review of the latter approaches revising crisp and fuzzy approaches for dealing with NN in RSs. Eventually it points out as a future research the management of NN in other recommendation scenarios as group RSs.

show abstract

“…The best value for d 1 has been determined empirically [14], concluding that for d 1 = 1 the proposal obtains an optimal performance.…”

Section: Noise Detection Phasementioning

confidence: 99%

Section: Martin@ujaenesmentioning

confidence: 99%

Section: A Fuzzy Approach To Detect Noisy Ratingsmentioning

confidence: 99%

“…Several experiments [14] were performed to determine the most suitable value for the threshold, d 2 , determining that d 2 = 1 is the best value.…”

Section: Noise Detection Phasementioning

confidence: 99%

See 2 more Smart Citations

Managing Natural Noise in Recommender Systems

Martínez

Castro

Toledo

2016

Theory and Practice of Natural Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…If users are closely connected or linked to each other in their social circle, there is a high probability that they have similar interests and interact with each other actively [11]. However the elicitation of customers preferences is not always precise either correct, because of external factors such as human errors, uncertainty and vagueness proper of human beings and so on [12]. Recommendations for item cold-start are given through comparing the properties of a new items to the properties of those items that are known to be of liked by some kind users.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy prototype classifier based on items and its application in recommender system

Mei¹,

Gong²,

Li³

2017

IJCIS

View full text Add to dashboard Cite

Currently, recommender systems (RS) are incorporating implicit information from social circle of the Internet. The implicit social information in human mind is not easy to reflect in appropriate decision making techniques. This paper consists of 2 contributions. First, we develop an item-based prototype classifier (IPC) in which a prototype represents a social circle's preferences as a pattern classification technique. We assume the social circle which distinguishes with others by the items their members like. The prototype structure of the classifier is defined by two2-dimensional matrices. We use information gain and OWA aggregator to construct a feature space. The item-based classifier assigns a new item to some prototypes with different prototypicalities. We reform a typical data set-Iris data set in UCI Machine Learning Repository to verify our fuzzy prototype classifier. The second proposition of this paper is to give the application of IPC in recommender system to solve new item cold-start problems. We modify the dataset of MovieLens to perform experimental demonstrations of the proposed ideas.

show abstract

A fuzzy clustering‐based denoising model for evaluating uncertainty in collaborative filtering recommender systems

Zhu

Han

Gou

et al. 2018

Asso for Info Science & Tech

View full text Add to dashboard Cite

Recommender systems are effective in predicting the most suitable products for users, such as movies and books. To facilitate personalized recommendations, the quality of item ratings should be guaranteed. However, a few ratings might not be accurate enough due to the uncertainty of user behavior and are referred to as natural noise. In this article, we present a novel fuzzy clustering‐based method for detecting noisy ratings. The entropy of a subset of the original ratings dataset is used to indicate the data‐driven uncertainty, and evaluation metrics are adopted to represent the prediction‐driven uncertainty. After the repetition of resampling and the execution of a recommendation algorithm, the entropy and evaluation metrics vectors are obtained and are empirically categorized to identify the proportion of the potential noise. Then, the fuzzy C‐means‐based denoising (FCMD) algorithm is performed to verify the natural noise under the assumption that natural noise is primarily the result of the exceptional behavior of users. Finally, a case study is performed using two real‐world datasets. The experimental results show that our proposal outperforms previous proposals and has an advantage in dealing with natural noise.

show abstract

A fuzzy model for managing natural noise in recommender systems

Cited by 51 publications

References 38 publications

Managing Natural Noise in Recommender Systems

Managing Natural Noise in Recommender Systems

Fuzzy prototype classifier based on items and its application in recommender system

A fuzzy clustering‐based denoising model for evaluating uncertainty in collaborative filtering recommender systems

Contact Info

Product

Resources

About