Track sequence clustering method based on deep learning

Abstract

The invention relates to the field of data mining, in particular to a track sequence clustering method based on deep learning. The method comprises the following steps: step 1, a pre-training layer: learning low-dimensional feature representation of track data by using a sequence-to-sequence auto-encoder model; step 2, an initial clustering layer: executing a K-Means clustering algorithm for multiple times on track feature representation obtained by the pre-training layer, and selecting a clustering center in optimal clustering results as an initial clustering center; step 3, a joint training optimization layer: combining track clustering and a depth feature extraction method, proposing an optimization loss function combining reconstruction errors and clustering errors of the sequence-to-sequence auto-encoder model, and mapping the track feature representation to a feature space more suitable for clustering.

Description

technical field [0001] The invention relates to the field of data mining, in particular to a trajectory sequence clustering method based on deep learning. Background technique [0002] The similarity measurement between trajectories is the basis of the spatio-temporal trajectory clustering method. Most trajectory clustering algorithms split the complete trajectory into segments or groups, and use point matching or self-defined strategies to compare the similarity between trajectories. Then use a widely popular clustering algorithm to gather similar trajectory objects into clusters. The accuracy of this clustering method needs to be improved. The development of deep learning makes it possible to learn feature representations of complex input sequences, which can be applied to the field of trajectory clustering to learn nonlinear feature representations that are more suitable for clustering and obtain clustering results with higher accuracy. Contents of the invention [000...

AI Technical Summary

Problems solved by technology

[0002] The similarity measure between trajectories is the basis of spatio-temporal trajectory clustering methods. Most trajectory clustering algorithms split complete trajectories into segments or groups, and use point matching or self-defined strategies to compare the similarity between trajectories. Then use a widely popular clustering algorithm to gather similar trajectory objects into clusters. The accuracy of this clustering method needs to be improved

Images