Deep Modeling of Group Preferences for Group-Based Recommendation
Authors: Liang Hu, Jian Cao, Guandong Xu, Longbing Cao, Zhiping Gu, Wei Cao
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Finally, the experiments conducted on a real-world dataset prove the superiority of our deep model over other state-of-the-art methods. |
| Researcher Affiliation | Academia | 1Shanghai Jiaotong University, 2University of Technology Sydney, 3Shanghai Technical Institute of Electronics & Information |
| Pseudocode | No | The paper does not contain any clearly labeled pseudocode or algorithm blocks. It describes the model and inference using mathematical equations and textual explanations. |
| Open Source Code | No | The paper does not provide any statement about making source code publicly available or a link to a code repository. |
| Open Datasets | Yes | CAMRa2011 (Said et al. 2011) released a real-world dataset containing the movie watching records of households and the ratings on each watched movie given by some group members. |
| Dataset Splits | Yes | The dataset for track 1 of CAMRa2011 has 290 households with a total of 602 users who gave ratings (on a scale 1~100) over 7,740 movies. This dataset has been partitioned into a training set and an evaluation set. The training set contains 145,069 ratings given by those 602 members, and 114,783 movie choice records from the view of 290 groups. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., GPU, CPU models, memory) used to run the experiments. |
| Software Dependencies | No | The paper describes the use of RBMs and DBNs as building blocks but does not specify any software libraries or their version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | In the experiments, we tune the hyper parameters for each model, e.g. the dimensionality of latent features and the regularization parameters, by cross validation. Specially, we set 𝛽= 1 and 𝛼= 0.5 (cf. Eq. (13)) for OCRBM and DLGR when no strategy is used, and we set 𝛼= 1 and 𝑓(𝑔, 𝑖) = 1 [1 + log 𝑠(𝑔, 𝑖)] when a strategy 𝑠( ) is used. Also, we used similar settings for the weights of OCMF. |