Intent-aware Recommendation via Disentangled Graph Contrastive Learning

Authors: Yuling Wang, Xiao Wang, Xiangzhou Huang, Yanhua Yu, Haoyang Li, Mengdi Zhang, Zirui Guo, Wei Wu

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments are conducted on three datasets, which demonstrates the effectiveness of our proposed IDCL. Futher analysis shows that the learned intent representations and behavior distributions are interpretable.
Researcher Affiliation Collaboration 1Beijing University of Posts and Telecommunications 2Beihang University 3Meituan 4Tsinghua University
Pseudocode No The paper describes the model architecture and components using equations and descriptive text, but it does not include any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper mentions PyTorch as the implementation framework (Footnote 2: "We implement our model based on Pytorch.2 https://pytorch.org/") and Rec Bole for baselines (Footnote 3: "We implement all the baselines with the unified opensource of recommendation algorithms, i.e., Rec Bole 3 [Zhao et al., 2020].3 https://github.com/RUCAIBox/Rec Bole"). However, it does not provide an explicit statement or link to the source code for the IDCL methodology itself.
Open Datasets Yes We conduct our experiments on three real-world datasets. In detail, for two Movie Lens datasets with different scales (i.e., ML-100k, ML-1M) [Harper and Konstan, 2015]
Dataset Splits Yes We split all users into training/validation/test sets as Multi VAE
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU models, CPU types, memory) used for running the experiments. It only mentions the software framework.
Software Dependencies No The paper states: "We implement our model based on Pytorch." However, it does not specify the version number of PyTorch or any other software dependencies required to replicate the experiments.
Experiment Setup Yes We turn the hyper-parameters in validation set using random search, and the search space of some important hyperparameters are: K {6, 8, 10, 12, 14, 16}, d [20, 40]. ... The Adam optimizerfor mini-batch gradient descent is applied to train all models.