Content-Aware Point of Interest Recommendation on Location-Based Social Networks
Authors: Huiji Gao, Jiliang Tang, Xia Hu, Huan Liu
AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we evaluate the performance of our proposed framework CAPRF for POI recommendation. In particular, we evaluate the following: (1) how the proposed framework fares in comparison with state-of-the-art recommender systems; and (2) how different kinds of content information perform in the POI recommendation task. Before we delve into experiment details, we first discuss an LBSN dataset and evaluation metrics. |
| Researcher Affiliation | Academia | Huiji Gao, Jiliang Tang, Xia Hu, and Huan Liu Computer Science and Engineering Arizona State University {Huiji.Gao, Jiliang.Tang, Xia.Hu, Huan.Liu}@asu.edu |
| Pseudocode | Yes | Algorithm 1 Learning Algorithm of the Proposed Model |
| Open Source Code | No | The paper does not provide any statement about releasing open-source code or a link to a code repository for their methodology. |
| Open Datasets | No | We choose Foursquare, one of the most popular locationbased social networking sites, to study the content information on LBSNs. We collect users whose Foursquare profiles indicate their hometown as California state. We then obtain their corresponding check-in tweets with the same crawling strategy as proposed in (Scellato et al. 2011; Gao, Tang, and Liu 2012), and collect check-ins that happened in California state. |
| Dataset Splits | No | For each individual user in the check-in matrix, we randomly mark off20% of all POIs that he has checked-in for testing. The rest of the observed user-POI pairs are used as training data for POI recommendation. The random selection is conducted 5 times individually, and we report the average results. ... All the parameters in this paper are set through cross-validation. |
| Hardware Specification | No | The paper does not provide any specific hardware details such as GPU models, CPU types, or memory used for running the experiments. |
| Software Dependencies | No | The paper mentions using a sentiment classification method and the MPQA Subjectivity Lexicon, but does not provide specific version numbers for any software, libraries, or programming languages used in their implementation. |
| Experiment Setup | Yes | For the proposed method, the experimental results use d=20 dimensions to represent the latent features, the parameters {η, λ1, λ2, δ, α} are set to {0.3, 0.1, 0.1, 0.8, 0.1}. |