Model‐based evaluation for online shopping platform with probabilistic double hierarchy linguistic CODAS method

Lei, Fan; Wei, Guiwu; Chen, Xudong

doi:10.1002/int.22514

Cited by 63 publications

(15 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first category does not utilize deep neural network components, with the matrix factorization (MF)‐based model being the most widely used method. This type of method usually predicts user preferences on items through evaluating the dot‐product values between the latent features of users and items 33 . Salakhutdinov et al 2,3 propose two models to provide effective recommendations for users, that is, probabilistic matrix factorization (PMF) and Bayesian PMF.…”

Section: Related Workmentioning

confidence: 99%

“…This type of method usually predicts user preferences on items through evaluating the dot-product values between the latent features of users and items. 33 Salakhutdinov et al 2,3 propose two models to provide effective recommendations for users, that is, probabilistic matrix factorization (PMF) and Bayesian PMF. PMF is a probabilistic linear model with Gaussian observation noise trained via maximizing the log-posterior on the user and item latent features utilizing the observation noise variance and prior variances.…”

Section: Traditional Recommendation Modelsmentioning

confidence: 99%

See 1 more Smart Citation

A two‐phase knowledge distillation model for graph convolutional network‐based recommendation

Huang

Lin

Gong

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Graph convolutional network (GCN)‐based recommendation has recently attracted significant attention in the recommender system community. Although current studies propose various GCNs to improve recommendation performance, existing methods suffer from two main limitations. First, user–item interaction data is generally sparse in practice, highlighting these methods' ineffectiveness in learning user and item feature representations. Second, they usually perform a dot‐product operation to model and calculate user preferences on items, leading to inaccurate user preference learning. To address these limitations, this study adopts a design idea that sharply differs from existing works. Specifically, we introduce the knowledge distillation concept into GCN‐based recommendation and propose a two‐phase knowledge distillation model (TKDM) improving recommendation performance. In Phase I, a self‐distillation method on a graph auto‐encoder learns the user and item feature representations. This auto‐encoder employs a simple two‐layer GCN as an encoder and a fully connected layer as a decoder. On this basis, in Phase II, a mutual‐distillation method on a fully connected layer is introduced to learn user preferences on items with triple‐based Bayesian personalized ranking. Extensive experiments on three real‐world data sets demonstrate that TKDM outperforms classic and state‐of‐the‐art methods related to GCN‐based recommendation problems.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Traditional Recommendation Modelsmentioning

confidence: 99%

A two‐phase knowledge distillation model for graph convolutional network‐based recommendation

Huang

Lin

Gong

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

show abstract

“…Considering that the probability sum of all PDHLEs in PDHLTS may not be 1, the following normalization formula is proposed. Definition 4 (Lei et al, 2021). Let…”

Section: { }mentioning

confidence: 99%

“…Recently, to solve this issue the probabilistic double hierarchy linguistic term set (PDHLTS) proposed by Gou, Xu, Liao, and Herrera (2021). and Lei, Wei, and Chen (2021) defined the probabilistic double hierarchy linguistic CODAS method. So in PDHLTS, we can use { }…”

Section: Introductionmentioning

confidence: 99%

PDHL-Edas Method for Multiple Attribute Group Decision Making and Its Application to 3d Printer Selection

Lei

Wei

Shen

et al. 2021

Technological and Economic Development of Economy

Self Cite

View full text Add to dashboard Cite

With the rapid development of 3D printing technology, 3D printers are manufactured based on the principle of 3D printing technology are more and more widely used in the manufacturing industry. Choosing high quality 3D printers for industrial production is of great significance to the economic growth of enterprises. In fact, it is difficult to select the most optimal 3D printers under a single and simple standard. Therefore, this paper establishes the probabilistic double hierarchy linguistic EDAS (PDHL-EDAS) method for the multiple attribute group decision making (MAGDM). Then the CRITIC model is introduced to derive objective weight and the cumulative prospect theory is leaded into obtain the cumulative weight of PDHLTS. In addition, what’s more, the PDHL-EDAS method is built and applied to the choice of high-quality 3D printer. Finally, compared with the available MAGDM methods under PDHLTS, the built method is proved to be scientific and effective.

show abstract

“…Both multiple attribute decision making (MADM) and multiple attribute group decision making (MAGDM) (Kim et al, 1999;Liao et al, 2021;Xu & Chen, 2007;Zhao et al, 2021b) refer to the process of selecting the optimal alternative among several alternatives, and these alternatives involve multiple identical attributes (Chen et al, 2012;Lei et al, 2021;Wei et al, 2021). In our daily life, we have to face many complex and uncertain questions that cannot be answered with full determined or absolute negation, so does the management decisionmaking process.…”

Section: Introductionmentioning

confidence: 99%

CPT-Todim Method for Interval Neutrosophic Magdm and Its Application to Third-Party Logistics Service Providers Selection

Zhang

Zhao

et al. 2021

Technological and Economic Development of Economy

Self Cite

View full text Add to dashboard Cite

The multiple attribute group decision making (MAGDM) has always been a concern in the research field. In this article, we establish the interval neutrosophic TODIM method based on cumulative prospect theory (CPT-IN-TODIM) for MAGDM issue. This new developed CPTIN-TODIM method has markedly superiority in describing decision maker’s psychological states, which utilizes the weight function to adjust weighting attributes distinguishing from the classical TODIM method. Then, this new developed method has been applied to select the third-party logistics service providers and been expound on the disparity with existing methods. Finally, the results of contrastive analysis indicate that this new developed method can lead to the appropriate conclusion and sticks out the differences between alternatives to provide clearer direction. Hence, the new developed CPT-IN-TODIM method is reliable and valid.

show abstract

Model‐based evaluation for online shopping platform with probabilistic double hierarchy linguistic CODAS method

Cited by 63 publications

References 40 publications

A two‐phase knowledge distillation model for graph convolutional network‐based recommendation

A two‐phase knowledge distillation model for graph convolutional network‐based recommendation

PDHL-Edas Method for Multiple Attribute Group Decision Making and Its Application to 3d Printer Selection

CPT-Todim Method for Interval Neutrosophic Magdm and Its Application to Third-Party Logistics Service Providers Selection

Contact Info

Product

Resources

About