Identifying Human Mobility via Trajectory Embeddings

Authors: Qiang Gao, Fan Zhou, Kunpeng Zhang, Goce Trajcevski, Xucheng Luo, Fengli Zhang

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments conducted on real-world datasets demonstrate that TULER achieves better accuracy than the existing methods. We now present our experiments, comparing TULER with several baseline methods on two public datasets.
Researcher Affiliation Academia University of Electronic Science and Technology of China, Chengdu, China University of Maryland, College park Northwestern University, Evanston
Pseudocode No The paper describes the LSTM and GRU models using mathematical equations but does not present any structured pseudocode or algorithm blocks.
Open Source Code Yes Source code, datasets and implementation details are available online at https://github.com/gcooq/TUL.
Open Datasets Yes To show the performance of TULER and the comparison with some existing methods, we conduct our experiments on two publicly available LBSN datasets: Gowalla and Brightkite [Cho et al., 2011].
Dataset Splits No Table 1 shows 'S/T: the number of trajectories for training and testing' (e.g., Gowalla 17,654/2,063), but does not explicitly mention a validation split or specific percentages for training, validation, and testing.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions general software components like RNN variants (LSTM, GRU) but does not provide specific version numbers for programming languages, libraries, or other software dependencies used in the experiments.
Experiment Setup Yes Table 2: Parameters used in TULER and baselines. Parameters We choose Possible Choices Dimensionality 250 100-300 Hidden size 300 250-1000 Learning rate 0.00095 0.00085-0.1 Dropout rate 0.5 0-1 Stacked TULER 2 2