用户轨迹识别作为一项重要的时空数据挖掘任务,广泛应用于基于位置的个性化服务推荐、行程规划、犯罪行为检测和目标跟踪等领域,但依然面临预测精度不高的问题,主要原因是轨迹数据低采样且稀疏、轨迹类别数量巨大等.针对上述问题提出了基于可拓展自注意力时空图卷积神经网络的用户轨迹识别模型(Expandable Self-At-tention Spatio-Temporal Graph Convolutional Neural Networks,ESAST-GCNN),该模型采用时空图卷积神经网络方式,深度挖掘时序特征与空间特征关系,并进行预测与拓展,结合自注意力机制获取用户轨迹特征向量内部相关性,最终根据该特征向量进行用户轨迹身份识别.在两个真实数据集上进行测试后发现,ESAST-GCNN相较于TULER-GRU(TUL via Embedding and RNN)在Geolife与Gowalla中准确率分别提高了13.95%、10.63%,实验结果表明ESAST-GCNN优于其他模型,识别效果更好,适用范围更广.
User Trajectory Identification Based on Expandable Self-Attention Spatio-Temporal Graph Convolutional Neural Networks
As an important spatio-temporal data mining task,user trajectory identification is widely used in the fields of location-based personalized service recommendation,itinerary planning,crime behavior detection,and target tracking.However,it still has low prediction accuracy,mainly due to low sampling and sparse trajectory data,and a huge number of trajectory categories.To fill the research gaps,a user trajectory identification model based on an expandable self-attention spatio-temporal graph convolutional neural network(ESAST-GCNN)is proposed,which adopts the spatio-temporal graph convolutional neural network to deeply mine the relationship between time sequence features and spatial features to predict and expand the sequence.This model combines the self-attention mechanism to obtain the internal correlation of user trajec-tory feature vectors and identify user trajectories.After testing on two real datasets,the results show that the accuracy of ES-AST-GCNN is improved by 13.95%and 10.63%in Geolife and Gowalla compared with TUL via Embedding and RNN(TULER-GRU),respectively.The experimental results illustrate that ESAST-GCNN is superior to other comparative mod-els,with better identification effect and wider applicability.
user trajectory identificationspatio-temporal graph convolutional neural networkself-attention mecha-nismdeep learningspace-time sequence
万方数据
维普
