Filters








34,021 Hits in 4.3 sec

From Motion Activity to Geo-Embeddings: Generating and Exploring Vector Representations of Locations, Traces and Visitors through Large-Scale Mobility Data

Alessandro Crivellari, Euro Beinat
2019 ISPRS International Journal of Geo-Information  
The algorithm consists of two steps, the trajectory pre-processing and the Word2vec-based model building.  ...  First, mobility traces are converted into sequences of locations that unfold in fixed time steps; then, a Skip-gram Word2vec model is used to construct the location embeddings.  ...  It contains 5.1 million trajectories, with an average trajectory length of 11.2 hours. The average traveled distance per hour is 13.4 km. .  ... 
doi:10.3390/ijgi8030134 fatcat:dao3yqt6vzegjlnvhppuxyucvy

Identifying Foreign Tourists' Nationality from Mobility Traces via LSTM Neural Network and Location Embeddings

Alessandro Crivellari, Euro Beinat
2019 Applied Sciences  
The problem is defined as a multinomial classification with a few tens of classes (nationalities) and sparse location-based trajectory data.  ...  The interest in human mobility analysis has increased with the rapid growth of positioning technology and motion tracking, leading to a variety of studies based on trajectory recordings.  ...  Similar to word embeddings in NLP [30] [31] [32] , we generate location embeddings θ i ∈ R d (d is the dimensionality of the embedding space) according to the motion behavior of people traveling over  ... 
doi:10.3390/app9142861 fatcat:5bjyxmetxfe5hpb64whzvecdzu

S2N2: An Interpretive Semantic Structure Attention Neural Network for Trajectory Classification

Canghong Jin, Ting Tao, Xianzhe Luo, Zemin Liu, Minghui Wu
2020 IEEE Access  
INDEX TERMS User classification, interpretive trajectory structure, human behavior understanding.  ...  sequential deep learning models, especially when trajectory semantic vectors are incorporated.  ...  Our approach shares some aspects with the abovementioned embedding techniques, but, unlike classical classification models, our model describes users' behaviors on the basis of their movement scenarios  ... 
doi:10.1109/access.2020.2982823 fatcat:j3gpjdh5hjhptazfatjqfo6fou

DETECT: Deep Trajectory Clustering for Mobility-Behavior Analysis [article]

Mingxuan Yue, Yaguang Li, Haoze Yang, Ritesh Ahuja, Yao-Yi Chiang, Cyrus Shahabi
2020 arXiv   pre-print
In addition, the expense of labeling massive trajectory data is a barrier to supervised learning models.  ...  To address these challenges, we propose an unsupervised neural approach for mobility behavior clustering, called the Deep Embedded TrajEctory ClusTering network (DETECT).  ...  Trajectories with similar mobility behavior exhibit various spatial and/or temporal range of movement.  ... 
arXiv:2003.01351v1 fatcat:wirelxw65rg77psfm6izezb6bm

Trace2trace—A Feasibility Study on Neural Machine Translation Applied to Human Motion Trajectories

Alessandro Crivellari, Euro Beinat
2020 Sensors  
The experiment is inserted in the background of tourist mobility analysis, with the goal of translating the motion behavior of tourists belonging to a specific nationality into the motion behavior of tourists  ...  The model adopted is based on the seq2seq approach and consists of an encoder–decoder architecture based on long short-term memory (LSTM) neural networks and neural embeddings.  ...  This baseline is useful for observing the degree of similarity between the motion behaviors of the two nationalities. • Highest similarity model.  ... 
doi:10.3390/s20123503 pmid:32575822 fatcat:scipwmrgcjgo5lmwqunkgjl3vm

Attention Based Sequence Learning Model for Travel Time Estimation

Zhong Wang, Hao Fu, Guiquan Liu, Xianwei Meng
2020 IEEE Access  
In this paper, we propose a novel attention based sequence learning model for travel time estimation of a path (ASTTE), that not only considers the real-world road network topology as multi-relational  ...  INDEX TERMS Travel time estimation, road network topology, multi-relational data  ...  In contrast, our model significantly outperforms other methods with a trajectory longer or shorter.  ... 
doi:10.1109/access.2020.3042673 fatcat:5vzpy5fbxrg6teg5nvov6nmucy

Detecting Suspects by Large-Scale Trajectory Patterns in the City

Cang-Hong Jin, Dong-Kai Chen, Fan-Wei Zhu, Ming-Hui Wu
2019 Mobile Information Systems  
We also propose two models to improve the identification performance, namely, the trajectory pattern model (TPM) and neural network-based model.  ...  The trajectory pattern model (TPM) offers a novel view to describe users' movement behaviors and generates more effective and universal features other than location and timestamp dimensions.  ...  We represent a user u with n trajectories as v u n [v u 1 , v u 2 , . . . , v u n ] , where a single trajectory embedding is concatenated to generate its vector space.  ... 
doi:10.1155/2019/1837594 fatcat:4qmtnwn6rzbkpj2jedtb5rzip4

Augmented Intention Model for Next-Location Prediction from Graphical Trajectory Context

Canghong Jin, Zhiwei Lin, Minghui Wu
2019 Wireless Communications and Mobile Computing  
The results demonstrate that the AI-RNN model outperforms other methods in terms of top-k accuracy, especially in scenarios with low similarity.  ...  Most existing approaches are sequential-model based and produce a prediction by mining behavior patterns.  ...  Figure 5 : 5 Heat Figure 6 : 6 Regularity of trajectories in different datasets. e x-axis is the length of trajectory and y axis-is the Jaccard similarity value. (a) Urban travel. (b) Foursquare.  ... 
doi:10.1155/2019/2860165 fatcat:7mkzr2ub7vhbvpodanxsimffiq

TrajGAIL: Generating Urban Vehicle Trajectories using Generative Adversarial Imitation Learning [article]

Seongjin Choi, Jiwon Kim, Hwasoo Yeo
2021 arXiv   pre-print
similar to real vehicle trajectories with limited observations.  ...  The model is tested with both simulation and real-world datasets, and the results show that the proposed model obtained significant performance gains compared to existing models in sequence modeling.  ...  The disaggregated travel behaviors include user-centric travel experiences, namely, speed profile, link-to-link route choice behavior and travel time experienced by individual vehicles, as well as system-wide  ... 
arXiv:2007.14189v4 fatcat:svsumpjklncxdh62g2zwyhmvnq

LSTM-Based Deep Learning Model for Predicting Individual Mobility Traces of Short-Term Foreign Tourists

Alessandro Crivellari, Euro Beinat
2020 Sustainability  
The increasing availability of trajectory recordings has led to the mining of a massive amount of historical track data, allowing for a better understanding of travel behaviors by revealing meaningful  ...  place in the trajectory.  ...  A graphic exemplifying overview of the whole model, with a block of two LSTM layers, is illustrated in Figure 1 . vector, encoding input trajectories into sequences of embeddings that are subsequently  ... 
doi:10.3390/su12010349 fatcat:pcxpqgjv7zauvo6anhu3wh62ce

PG^2Net: Personalized and Group Preferences Guided Network for Next Place Prediction [article]

Huifeng Li, Bin Wang, Fan Xia, Xi Zhai, Sulei Zhu, Yanyan Xu
2021 arXiv   pre-print
Predicting the next place to visit is a key in human mobility behavior modeling, which plays a significant role in various fields, such as epidemic control, urban planning, traffic management, and travel  ...  We adopt a graph embedding method to map users' trajectory into a hidden space, capturing their sequential relation.  ...  Researchers also notice that the users' periodical behavior observed from long-term history trajectory plays a critical role in the decision of next travel [15, 16] .  ... 
arXiv:2110.08266v1 fatcat:ea3pz4t4tvbq3aw5hahb6yydfm

Deep Learning based Urban Vehicle Trajectory Analytics [article]

Seongjin Choi
2021 arXiv   pre-print
In this study, we propose various novel models for urban vehicle trajectory analytics using deep learning.  ...  The urban vehicle trajectory analytics offers unprecedented opportunities to understand vehicle movement patterns in urban traffic networks including both user-centric travel experiences and system-wide  ...  They used a dynamic clustering algorithm with Wasserstein distance to make clusters of link pairs with similar travel time distribution.  ... 
arXiv:2111.07489v1 fatcat:zanf5aj7unfb5joey3f7lzhtbm

Tortuosity Entropy: a measure of spatial complexity of behavioral changes in animal movement data [article]

Xiaofeng Liu, Ning Xu, Aimin Jiang
2014 arXiv   pre-print
behavioral change in animal movement data in a fine scale.  ...  We test the algorithm on both simulated trajectories and real trajectories and show that both mixed segments in synthetic data and different phases in real movement data are identified accurately.  ...  Test on simulated trajectory data We first simulate an oriented travel trajectory with a consistent bias in a given preferred direction γ , which is modeled using biased random walks (BRWs) for a long  ... 
arXiv:1312.5231v2 fatcat:2ybpyjp2fvd55ieaah27mxokcq

Vehicle Identity Recovery for Automatic Number Plate Recognition Data via Heterogeneous Network Embedding

Yixian Chen, Zhaocheng He
2020 Sustainability  
The proposed method integrates the vehicle group entities and context relations into the THIN for capturing the spatiotemporal relationships in vehicle travel and adopts a holographic embeddings model  ...  Further, we compare it with other embedding methods and the results support the superiority of holographic embeddings.  ...  the information from vehicles with similar travel trajectories decreases.  ... 
doi:10.3390/su12083074 fatcat:wihxxbm2abbqtjhizxowc4uplm

Learning GPS Point Representations to Detect Anomalous Bus Trajectories

Michael Cruz, Luciano Barbosa
2020 IEEE Access  
To deal with that, previous approaches [3] , [5] have proposed to apply a classifier to model the regular behavior of bus trajectories by predicting their route line, and considering trajectories with  ...  The STOD embedding encodes relevant aspects regarding temporal and spatial behavior of the bus trajectory.  ... 
doi:10.1109/access.2020.3046912 fatcat:jcqutsp7mfexpm4enhcpjpmgc4
« Previous Showing results 1 — 15 out of 34,021 results