Related Data for COMET: Convolutional Dimension Interaction for Collaborative Filtering (doi:10.21979/N9/TO2HBX)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

(external link)

Document Description

Citation

Title:

Related Data for COMET: Convolutional Dimension Interaction for Collaborative Filtering

Identification Number:

doi:10.21979/N9/TO2HBX

Distributor:

DR-NTU (Data)

Date of Distribution:

2022-04-18

Version:

2

Bibliographic Citation:

Lin, Zhuoyi, 2022, "Related Data for COMET: Convolutional Dimension Interaction for Collaborative Filtering", https://doi.org/10.21979/N9/TO2HBX, DR-NTU (Data), V2

Study Description

Citation

Title:

Related Data for COMET: Convolutional Dimension Interaction for Collaborative Filtering

Identification Number:

doi:10.21979/N9/TO2HBX

Authoring Entity:

Lin, Zhuoyi (Nanyang Technological University)

Other identifications and acknowledgements:

Kwoh Chee Keong

Software used in Production:

python

Distributor:

DR-NTU (Data)

Access Authority:

Lin, Zhuoyi

Depositor:

Lin, Zhuoyi

Date of Deposit:

2022-04-18

Holdings Information:

https://doi.org/10.21979/N9/TO2HBX

Study Scope

Keywords:

Computer and Information Science, Computer and Information Science, recommendation, amazon dataset, yelp

Abstract:

Latent factor models play a dominant role among recommendation techniques. However, most of the existing latent factor models assume both historical interactions and embedding dimensions are independent of each other, and thus regrettably ignore the high-order interaction information among historical interactions and embedding dimensions. In this paper, we propose a novel latent factor model called COMET (COnvolutional diMEnsion inTeraction), which simultaneously model the high-order interaction patterns among historical interactions and embedding dimensions. To be specific, COMET stacks the embeddings of historical interactions horizontally at first, which results in two "embedding maps". In this way, internal interactions and dimensional interactions can be exploited by convolutional neural networks with kernels of different sizes simultaneously. A fully-connected multi-layer perceptron is then applied to obtain two interaction vectors. Lastly, the representations of users and items are enriched by the learnt interaction vectors, which can further be used to produce the final prediction. Extensive experiments and ablation studies on various public implicit feedback datasets clearly demonstrate the effectiveness and the rationality of our proposed method.

Kind of Data:

observation ratings

Methodology and Processing

Sources Statement

Data Access

Other Study Description Materials

Related Publications

Citation

Identification Number:

arXiv:2007.14129

Bibliographic Citation:

Lin, Z., Feng, L., Guo, X., Zhang, Y., Yin, R., Kwoh, C. K., & Xu, C. (2020). COMET: Convolutional Dimension Interaction for Collaborative Filtering. arXiv preprint arXiv:2007.14129.

Other Study-Related Materials

Label:

comet.rar

Notes:

application/x-rar-compressed

Other Study-Related Materials

Label:

ml-1m.test.negative

Notes:

application/octet-stream

Other Study-Related Materials

Label:

ml-1m.test.rating

Notes:

application/octet-stream

Other Study-Related Materials

Label:

ml-1m.train.rating

Notes:

application/octet-stream