Embedding geographic information for anomalous trajectory detection
- PDF / 1,311,501 Bytes
- 21 Pages / 439.642 x 666.49 pts Page_size
- 79 Downloads / 286 Views
Embedding geographic information for anomalous trajectory detection Ding Xiao1 · Li Song2 · Ruijia Wang1 · Xiaotian Han1 · Yanan Cai3 · Chuan Shi1 Received: 4 April 2019 / Revised: 10 December 2019 / Accepted: 12 March 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Anomalous trajectory detection is a crucial task in trajectory mining fields. Traditional anomalous trajectory detection methods mainly focus on the differences of a new trajectory and the historical trajectory with density and isolation techniques, which may suffer from the following two disadvantages. (1) They cannot capture the sequential information of the trajectory well. (2) They cannot make use of the common information of the trajectory points. To overcome the above shortcomings, we propose a novel method called Anomalous Trajectory Detection using Recurrent Neural Network (AT D − RN N ) which characterizes the trajectory with the learned trajectory embedding. The trajectory embedding can capture the sequential information of the trajectory and depict the internal characteristics between abnormal and normal trajectories. In order to learn the high-quality trajectory embedding, we further propose an attention mechanism to aggregate the long sequential information. Furthermore, to alleviate the data sparsity problem, we augment the datasets between a source and a destination by taking the relevant trajectories into consideration simultaneously. Extensive experiments on real-world datasets validate the effectiveness of our proposed methods. Keywords Anomalous trajectory detection · Attention mechanism · Trajectory embedding · Recurrent neural network
1 Introduction With the proliferation of global positioning system (GPS) based equipment, massive spatial trajectory data has been generated. The trajectory data represents the mobility of a diversity of moving objects and contains valuable information concerning both Service Providers (SPs) and the customers. A large number of trajectory data mining tasks [53], including This article belongs to the Topical Collection: Special Issue on Application-Driven Knowledge Acquisition Guest Editors: Xue Li, Sen Wang, and Bohan Li Chuan Shi
[email protected]
Extended author information available on the last page of the article.
World Wide Web
map matching, trajectory compression, trajectory classification and anomalous trajectory detection, have been widely researched. Anomalous trajectory detection plays an important role in the fields of trajectory data mining. For example, anomalous trajectory detection can enhance the quality of taxi services impressively, since the greedy taxi drivers who overcharge passengers could be detected with anomalous trajectory detection techniques. Therefore, it allows the taxi companies to monitor the movements of all the taxis and to identify the dishonest drivers who tend to take routes longer than usual. A toy example of anomalous and normal trajectory between a source (S) and destination (D) (SD − P air) can be seen in Figure 1
Data Loading...
