Destination prediction by sub-trajectory synthesis and privacy protection against such prediction

Xue, Andy Yuan; Zhang, Rui; Zheng, Yu; Xie, Xing; Huang, Jin; Xu, Zhenghua

doi:10.1109/icde.2013.6544830

Cited by 207 publications

(153 citation statements)

References 21 publications

Supporting

Mentioning

151

Contrasting

Unclassified

Order By: Relevance

“…We evaluate the prediction accuracy by the proposed framework with various scoring functions, called High Frequency First(HFF), Near Location First(NLF), Near Time First(NTF), and Combination(Combine). Additionally, we compare the performance with the existing methods [9,10]. In [10], the authors predict the destination by computing the posterior probability for any given query subtrajectory(Sub-Trajectory Synthesis).…”

Section: Evaluation Of Next Region Predictionmentioning

confidence: 99%

“…Additionally, we compare the performance with the existing methods [9,10]. In [10], the authors predict the destination by computing the posterior probability for any given query subtrajectory(Sub-Trajectory Synthesis). In [9], they explore the trajectory patterns and construct the decision tree to predict the next stay location(Trajectory Pattern).…”

Section: Evaluation Of Next Region Predictionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…Markov model has been widely applied in predicting destinations for a specific individual as well [1,10]. In [10], the author uses a Markov model to offline prepare the probabilities needed to efficiently compute the posterior probability for any given query trajectory online. Some existing works use the external information to predict destination [6,12], these external information such as the distributions of different districts (ground cover), of traveling time, of trajectories length, the accident reports, road condition, and driving habits often enhance the prediction accuracy.…”

Section: Related Workmentioning

confidence: 99%

“…Given a set of trajectories, prior works have studied the location prediction problem in which given the users current location, the problem is to predict the next location or the location at a specific time. In [7,10], the authors first map the user's historical trajectories into the regions by grid-cell system and estimate the transition probability between the mapped regions. Then, according to the user's current location to predict the future movement based on the probability model.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Mining Top-K Relevant Stay Regions from Historical Trajectories

Lin

Lai

Lei

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. With increasingly prevalent mobile positioning devices, such as GPS loggers, smart phones, and GPS navigation devices, a huge amount of trajectories data is collected. Users are able to obtain the various location-based services by uploading their trajectories. In this paper, we address that a user's movement behavior is able to discover by their similar shape trajectories and resulted in some regions frequently stay in common, called relevant stay regions. Once a set of stay regions discovered, we can predict the next region where the user intends to go and provide location-based information of the next stay in advance, such as traffic status, targeted advertises, sightseeing recommendations, and so on. Prior works have elaborated on discovering stay region from the whole crowd trajectories and then exploring the relations between the regions to describe the movement patterns for location prediction. However, the trajectories pass the same region may not have the similar movement behavior. Thus, we propose a framework to discover stay regions relevant to the specific movement behavior and then applied in location prediction, called Region Modeling and Mobility Prediction. The proposed framework includes two modules: region modeling and mobility prediction. In the region modeling module, we develop shape clustering method to group the similar trajectories from historical data and then explore the stay region model from trajectory clusters. Based on the discovered region model, the mobility prediction module provide a cluster selection algorithm and several prediction strategies to generate the topk relevant stay regions. Experiments results on real datasets demonstrate the effectiveness and accuracy of our proposed model on detecting next stay region, comparing with other baseline methods.

show abstract

Section: Evaluation Of Next Region Predictionmentioning

confidence: 99%

Section: Evaluation Of Next Region Predictionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Mining Top-K Relevant Stay Regions from Historical Trajectories

Lin

Lai

Lei

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

Dispersed dummy selection approach for location‐based services to preempt user‐profiling

Manju

Sumathy

2021

Concurrency and Computation

View full text Add to dashboard Cite

Location-based services (LBS) provide specific personalized services based on the location information provided by the user. Disclosing the users' private data is a challenging issue prevailing in LBS. Users become victims of online user-profile disclosure risks when they continuously utilize the services and attempt to disclose private data. The proposed dispersed dummy selection-based approaches such as dispersed dummy selection including actual location (DDSIA) and dispersed dummy selection excluding actual location (DDSEA) avert user-profiling issue. The dummy locations are chosen based on the auxiliary information of the location, user, and physical dispersion. The auxiliary information about the location is fetched from the local fog servers, and auxiliary information about the user is stored and retrieved from the cloud storage. An anonymous circular area is considered to ensure the dummy locations are dispersed, and the user-location correlations get dissipated. The detailed analysis of the proposed approaches is explored with the real-world map data, and its effectiveness is verified with simulations using Matlab. In addition, the qualitative analysis tests the resistance of the dummies against human observations. The results obtained substantiate that the proposed approaches are efficient in terms of dispersion degree and resistance to human observations that avert user profiling.

show abstract

A survey of evaluation methods for personal route and destination prediction from mobility traces

Stegmann

Žliobaitė

Tolvanen

et al. 2017

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Personal mobility data can nowadays be easily collected by personal mobile phones and used for analytical modeling. To assist in such an analysis, a variety of computational approaches have been developed. The goal is to extract mobility patterns in order to provide traveling assistance, information, recommendations or on‐demand services. While various computational techniques are being developed, research literature on destination and route prediction lacks consistency in evaluation methods for such approaches. This study presents a review and categorization of evaluation criteria and terminology used in assessing the performance of such methods. The review is complemented by experimental analysis of selected evaluation criteria, to highlight the nuances existing between the evaluation measures. The experimental study uses previously unpublished mobility data of 15 users collected over a period of 6 months in Helsinki metropolitan area in Finland. The article is primarily intended for researchers developing approaches for personalized mobility analysis, as well as a guideline for practitioners to select criteria when assessing and selecting between computational approaches. Our main recommendation is to consider user‐specific accuracy measures in addition to averaged aggregates, as well as to take into consideration that for many users accuracy does not saturate fast and the performance keeps evolving over time. Therefore, we recommend using time‐weighted measures. WIREs Data Mining Knowl Discov 2018, 8:e1237. doi: 10.1002/widm.1237 This article is categorized under: Algorithmic Development > Spatial and Temporal Data Mining Application Areas > Society and Culture Application Areas > Industry Specific Applications

show abstract

Destination prediction by sub-trajectory synthesis and privacy protection against such prediction

Cited by 207 publications

References 21 publications

Mining Top-K Relevant Stay Regions from Historical Trajectories

Mining Top-K Relevant Stay Regions from Historical Trajectories

Dispersed dummy selection approach for location‐based services to preempt user‐profiling

A survey of evaluation methods for personal route and destination prediction from mobility traces

Contact Info

Product

Resources

About