Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Personalized Route Recommendation Based on User Habits for Vehicle Navigation

Yinuo Huang, Xin Jin, Miao Fan, Xunwei Yang, Fangliang Jiang

TL;DR

This paper proposes a Deep-Cross-Recurrent (DCR) learning model aimed at learning route sorting scores and offering customized route recommendations based on user historical navigation data and effectively captures recommended navigation routes and user preferences by integrating DCN-v2 and LSTM.

Abstract

Navigation route recommendation is one of the important functions of intelligent transportation. However, users frequently deviate from recommended routes for various reasons, with personalization being a key problem in the field of research. This paper introduces a personalized route recommendation method based on user historical navigation data. First, we formulate route sorting as a pointwise problem based on a large set of pertinent features. Second, we construct route features and user profiles to establish a comprehensive feature dataset. Furthermore, we propose a Deep-Cross-Recurrent (DCR) learning model aimed at learning route sorting scores and offering customized route recommendations. This approach effectively captures recommended navigation routes and user preferences by integrating DCN-v2 and LSTM. In offline evaluations, our method compared with the minimum ETA (estimated time of arrival), LightGBM, and DCN-v2 indicated 8.72%, 2.19%, and 0.9% reduction in the mean inconsistency rate respectively, demonstrating significant improvements in recommendation accuracy.

Personalized Route Recommendation Based on User Habits for Vehicle Navigation

TL;DR

This paper proposes a Deep-Cross-Recurrent (DCR) learning model aimed at learning route sorting scores and offering customized route recommendations based on user historical navigation data and effectively captures recommended navigation routes and user preferences by integrating DCN-v2 and LSTM.

Abstract

Navigation route recommendation is one of the important functions of intelligent transportation. However, users frequently deviate from recommended routes for various reasons, with personalization being a key problem in the field of research. This paper introduces a personalized route recommendation method based on user historical navigation data. First, we formulate route sorting as a pointwise problem based on a large set of pertinent features. Second, we construct route features and user profiles to establish a comprehensive feature dataset. Furthermore, we propose a Deep-Cross-Recurrent (DCR) learning model aimed at learning route sorting scores and offering customized route recommendations. This approach effectively captures recommended navigation routes and user preferences by integrating DCN-v2 and LSTM. In offline evaluations, our method compared with the minimum ETA (estimated time of arrival), LightGBM, and DCN-v2 indicated 8.72%, 2.19%, and 0.9% reduction in the mean inconsistency rate respectively, demonstrating significant improvements in recommendation accuracy.
Paper Structure (13 sections, 4 equations, 5 figures, 2 tables)

This paper contains 13 sections, 4 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Feature Extraction
  5. Route features.
  6. User profiles.
  7. Route Rank Model
  8. Experiments
  9. Datasets
  10. Evaluation Metrics
  11. Competing method
  12. Comparison results
  13. Conclusion

Figures (5)

  • Figure 1: Misalignment between the user trajectory and navigation route. The user diverged from the navigation's recommended route, opting instead for a scenic route along the riverbank.
  • Figure 2: Landscape information extraction. Convert the route into a grid sequence and then statistic the number or proportion of grids that include water systems data and green spaces data.
  • Figure 3: Visual cluster results of T-SNE. The K-Means method is used to cluster data and visualize them following dimensionality reduction by t-SNE.
  • Figure 4: Structure of DCR. The DCN-v2 network is on the left and the recurrent network is on the right.
  • Figure 5: The result comparison of different distance types. The test dataset is classified into three distance types, and then the effects of different models on each set of data are compared.