Sensitivity of Attributes on the Performance of Attribute-Aware Collaborative Filtering

Tso, Karen H. L.; Schmidt-Thierne, Lars

doi:10.1007/3-540-35978-8_32

Cited by 1 publication

(4 citation statements)

References 7 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From our previous findings [39], we have shown quality of attributes does have a great impact on the performance of the algorithms and that by adding useful attributes to the system, the performance of attribute-aware RS algorithms reaches its peak when the informativeness of attributes attain maximum. On this basis, we can examine this property further by using synthetic data with varying attribute informativeness.…”

Section: Results With Synthetic Data With Attributesmentioning

confidence: 89%

“…Thus, special effort is needed for selecting an assortment of useful attributes. Otherwise, attributes could act as noises and thus lead to poor recommendations [39]. Therefore, it is desirable to use synthetic data to evaluate RS algorithms, especially attribute-aware RS algorithms, before using real-life datasets.…”

Section: Data With Attributesmentioning

confidence: 99%

“…We have done empirical analysis to investigate this matter by looking at an augmented sets of movie attributes from the Amazon taxonomy of 1074 different genres/classes provided by Ziegler [38]. Ten other subsets of ml data using two quantitative measures to select informative attributes are used in this experiment [39]. The first measure considers the total number of attribute occurrences (attribute frequency) and second one considers the χ 2 measure between item occurrences and attributes.…”

Section: B Experiments On the Effect Of Attributesmentioning

confidence: 99%

“…The first measure considers the total number of attribute occurrences (attribute frequency) and second one considers the χ 2 measure between item occurrences and attributes. Confer [39] for further information on the experiments. We expected that good attribute-aware algorithms are sensitive to attributes as this implies attributes could influence the performance of recommender systems.…”

Section: B Experiments On the Effect Of Attributesmentioning

confidence: 99%

See 3 more Smart Citations

Empirical Analysis of Attribute-Aware Recommender System Algorithms Using Synthetic Data

Tso¹,

Schmidt-Thieme²

2006

JCP

Self Cite

View full text Add to dashboard Cite

Abstract-As the amount of online shoppers grows rapidly, the need of recommender systems for e-commerce sites are demanding, especially when the number of users and products being offered online continues to increase dramatically. There have been many ongoing researches on recommender systems and in investigating recommendation algorithms that could optimize the recommendation quality. However, adequate and public datasets of users and products have always been demanding to better evaluate recommender system algorithms. Yet, the amount of public data, especially data containing adequate content information (attributes) is limited. When evaluating recommendation algorithms, it is important to observe the behavior of the algorithm as the characteristic of data varies. Synthetic data would allow the application of systematic changes on the data which cannot be done with real-life data. Although studies on synthetic data for the use of recommender systems have been investigated, artificial data with attributes information are rarely looked into. In this paper, we review public and synthetic data that are applied in the field of recommender systems. A synthetic data generation methodology that considers attributes will also be discussed. Furthermore, we present empirical evaluations on existing attributeaware recommendation algorithms and other state-of-theart algorithms using real-life dataset as well as variable synthetic data to observe their behavior as the characteristic of data varies. In particular, the informativeness of attributes is being further investigated with both real-life datasets with augmented attributes sets as well as synthetic datasets with attributes. We have shown that a reasonably good overview of the behavior of attribute-aware algorithms can be obtained by using synthetic data compared to results done with real-life datasets.

show abstract

Section: Results With Synthetic Data With Attributesmentioning

confidence: 89%

Section: Data With Attributesmentioning

confidence: 99%