Contextualized Point-of-Interest Recommendation
Authors: Peng Han, Zhongxiao Li, Yong Liu, Peilin Zhao, Jing Li, Hao Wang, Shuo Shang
IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The results in our experiments show that our algorithm outperforms all the state-of-the-art methods. |
| Researcher Affiliation | Collaboration | 1University of Electronic Science and Technology of China 2King Abdullah University of Science and Technology 3Nanyang Technological University 4Tencent AI Lab 5Inception Institute of Artificial Intelligence |
| Pseudocode | Yes | The optimization procedure is summarized in Algorithm 1. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing source code or a direct link to a code repository for the described methodology. |
| Open Datasets | Yes | The Gowalla dataset1 contains user check-in data from February 2009 till October 2010 on the Gowalla social network. The dataset contains 18,737 users and 32,510 POIs. The total number of check-ins is 1,278,274. 1http://snap.stanford.edu/data/loc-gowalla.html The Yelp dataset2 contains 860,888 check-ins of 30,887 users and 18,995 POIs. They are from round 7 of the Yelp dataset challenge. 2https://www.yelp.com/dataset/challenge |
| Dataset Splits | Yes | The two experimental datasets are partitioned into training set, tuning set and test set. The check-ins of each user is splitted with ratio 70%, 20% and 10% for training, tuning and testing, respectively. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU models, CPU types, or memory specifications used for running the experiments. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x, etc.) that would be needed for replication. |
| Experiment Setup | Yes | The parameters in our model are sampled as follows: λ1 [0.05, 10], λ2 [0.05, 10], λ4 [10 6, 10 1], the user smoothing factor αuser [0, 1], the POI smoothing factor αpoi [0, 1], Gaussian kernel standard deviation σ [0.1, 10] and the number of clusters for spectral clustering G {20, 50, 100}. |