Trajectory-User Linking via Variational AutoEncoder

Authors: Fan Zhou, Qiang Gao, Goce Trajcevski, Kunpeng Zhang, Ting Zhong, Fengli Zhang

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 5 Evaluation In this section we present the evaluation of the benefits of TULVAE using three real-word GTSM datasets. To ease the reproduction of our results, we have made the source code of TULVAE publicly available1. Table 1: Data description: |U|: the number of users; |Tn|/|Te|: number of trajectories for training and testing; |C|: number of check-ins; R: average length of trajectories (before segmentation); Tr: range of the trajectory length Datasets: We conducted our experiments on three publicly available GTSM datasets: Gowalla2, Brightkite3 and Foursquare4. For Foursquare, we choose the most popular city New York. We randomly select |U| users and their corresponding trajectories from the datasets for evaluation for each dataset, we select two different numbers of those users (e.g., labels here) who generate varied trajectories for robustness check of model performance. Table 1 depicts the statistics of the three datasets. Baselines and Metrics: We compare TULVAE with several state-of-the-art approaches from the field of trajectory similarity measurement and deep learning based classification. We also implemented three hierarchical variations of TULER (TUL via Embedding and RNN) [Gao et al., 2017] namely: HTULER-L, HTULER-G and HTULER-B, respectively implemented with the hierarchical LSTM, GRU and Bi-directional LSTM but without variational inference. In our implementation, multivariate Gaussian distribution is used as the prior in TULVAE. The learning rate of all models is initialized with 0.001 and decays with rate of 0.9. The weight β (KL cost annealing) increases from 0.5 to 1; and the dropout rate is 0.5. We embed POIs in 250 dimensional vectors and used 300 units for classifier, 512 units for the encoder-decoder RNN and 100 units for latent variable z. Finally, the batch size is 64 for all RNN based models. The baselines used for benchmarking can be broadly categorized as: (a) Traditional approaches, including LDA (Linear Discriminant Analysis, with SVD as matrix solver), DT (Decision Tree), RF (Random Forest) and SVM (Support Support Vector Machine, with linear kernel), which are widely used for measuring mobility patterns and classifying trajectories in literatures [Zheng, 2015]. (b) RNN based TUL, including TULER-LSTM, TULERGRU, TULER-LSTM-S, TULER-GRU-S and Bi-TULER proposed in [Gao et al., 2017], which are the state-of-the-art methods for trajectory-user linking. We report the ACC@K, macro-P, macro-R and macro-F1 of all methods, which are common metrics in information retrieval area.
Researcher Affiliation Academia Fan Zhou1 , Qiang Gao1, Goce Trajcevski2, Kunpeng Zhang3, Ting Zhong1, Fengli Zhang1 1 School of Information and Software Engineering, University of Electronic Science and Technology of China. {fan.zhou, qianggao@std., zhongting@, fzhang@}uestc.edu.cn 2 Iowa State University, Ames. gocet25@iastate.edu 3 University of Maryland, College park. kpzhang@umd.edu
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes To ease the reproduction of our results, we have made the source code of TULVAE publicly available1. 1https://github.com/AI-World/IJCAI-TULVAE
Open Datasets Yes We conducted our experiments on three publicly available GTSM datasets: Gowalla2, Brightkite3 and Foursquare4. 2http://snap.stanford.edu/data/loc-gowalla.html 3http://snap.stanford.edu/data/loc-brightkite.html 4https://sites.google.com/site/yangdingqi/home
Dataset Splits No The paper mentions training and testing data splits in Table 1 ('|Tn|/|Te|: number of trajectories for training and testing') but does not specify a validation set or explicit split percentages for training/validation/test.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions architectural components like LSTM/GRU and models like LDA/SVM, but it does not specify software dependencies (e.g., deep learning frameworks, libraries) with version numbers.
Experiment Setup Yes The learning rate of all models is initialized with 0.001 and decays with rate of 0.9. The weight β (KL cost annealing) increases from 0.5 to 1; and the dropout rate is 0.5. We embed POIs in 250 dimensional vectors and used 300 units for classifier, 512 units for the encoder-decoder RNN and 100 units for latent variable z. Finally, the batch size is 64 for all RNN based models.