CAMO: A Collaborative Ranking Method for Content Based Recommendation
Authors: Chengwei Wang, Tengfei Zhou, Chen Chen, Tianlei Hu, Gang Chen5224-5231
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive empirical studies show that CAMO outperforms state-of-the-art methods in predicting users preferences. We conduct extensive experiments on benchmark datasets. |
| Researcher Affiliation | Academia | 1The Key Laboratory of Big Data Intelligent Computing of Zhejiang Province, China 2CAD & CG State Key Lab, Zhejiang University, China 3College of Computer Science and Technology, Zhejiang University, China |
| Pseudocode | Yes | Algorithm 1 Doc2Vec( vsent 1 , vsent 2 , . . . vsent Isent ) Algorithm 2 Alternate Gradient Method |
| Open Source Code | No | The paper states “The codes for the proposed CAMO and CRAE are implemented using Tensor Flow (Abadi et al. 2016)” but does not provide an explicit statement about releasing its own source code or a link to it. |
| Open Datasets | Yes | The benchmark datasets include: Cite ULike (Wang and Blei 2011), Mov Plot1M and Mov Plot10M (Liu et al. 2017). |
| Dataset Splits | No | The paper states: “We randomly select 80% of the observed data as training set and evaluate the models with the remaining 20%.” This describes a train/test split, but does not explicitly mention a separate validation split or its proportion. |
| Hardware Specification | Yes | All the compared methods are run on the same machine with i7-5930K CPU, 64GB RAM, and one TITAN Xp GPU. |
| Software Dependencies | No | The paper mentions that CAMO and CRAE are implemented using “Tensor Flow” but does not specify its version number or any other software dependencies with versions. |
| Experiment Setup | Yes | We set all the dimension hyperparameters (e.g. dimensions of the user, item and document latent vectors) of the considered models to the same value d. We call d model s dimension and set d = 256 for all compared methods by default. We set negative sampling ratio NS = 6. The default parameters for CAMO are set as follows: the number of topics Itopic is set to 10, the number of dynamic routing iterations r is set to 14, and the regularization parameter λ is set to 0.2. Other parameters of baseline methods are set to their default values. |