DeepMM: Deep Learning Based Map Matching with Data Augmentation

Li, Yong; Zhao, Kai; Zhao, Xu; Tong, Xia; Zhang, Jinglin; Jin, Depeng

doi:10.1109/tmc.2020.3043500

Cited by 21 publications

(14 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To make up for the shortcomings of traditional physical sensors, Shi et al [3][4][5][6] used social networks as social sensors to optimize the briefing content in the meteorological domain and provided online services for weather monitoring platforms. Feng et al 7 completed map matching in the latent space based on deep learning and enhanced the matching with the knowledge of mobile patterns. Yan et al 8 proposed an algorithm based on stretching and shrinking distance (SSDBA) for link prediction in social networks.…”

Section: Related Workmentioning

confidence: 99%

An explainable sentiment prediction model based on the portraits of users sharing representative opinions in social sensors

Wang

Zhu

et al. 2021

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

“Social sensors” refer to those who provide opinions through electronic communication channels such as social networks. There are two major issues in current models of sentiment analysis in social sensor networks. First, most existing models only analyzed the sentiment within the text but did not analyze the users, which led to the experimental results difficult to explain. Second, few studies extract the specific opinions of users. Only analyzing the emotional tendencies or aspect-level emotions of social users brings difficulties to the analysis of the opinion evolution in public emergencies. To resolve these issues, we propose an explainable sentiment prediction model based on the portraits of users sharing representative opinions in social sensors. Our model extracts the specific opinions of the user groups on the topics and fully considers the impacts of their diverse features on sentiment analysis. We conduct experiments on 51,853 tweets about the “COVID-19” collected from 1 May 2020 to 9 July 2020. We build users’ portraits from three aspects: attribute features, interest features, and emotional features. Six machine learning algorithms are used to predict emotional tendency based on users’ portraits. We analyze the influence of users’ features on the sentiment. The prediction accuracy of our model is 64.88%.

show abstract

Section: Related Workmentioning

confidence: 99%

An explainable sentiment prediction model based on the portraits of users sharing representative opinions in social sensors

Wang

Zhu

et al. 2021

International Journal of Distributed Sensor Networks

View full text Add to dashboard Cite

show abstract

“…Advanced algorithms use advanced techniques such as Kalman filter Jo et al (1996); Kim et al (2000), dynamic programming Zhu et al (2017) and Hidden Markov model Newson and Krumm (2009); Yang and Gidofalvi (2018); Feng et al (2020) in map-matching process. In Jo et al (1996); Kim et al (2000), they combined the geometric method with the Kalman filter to improve the matching accuracy.…”

Section: Advanced Algorithmsmentioning

confidence: 99%

“…In FMM, an upper-bounded origin-destination table is precomputed first to store all pairs of shorted paths under a specified length and then repeated routing queries are replaced with quick hash table searches. However, in Feng et al (2020), the authors point out two shortcomings in HMM based methods: they don't use historical trajectory in map-matching process and they are susceptible to noisy data.…”

Section: Advanced Algorithmsmentioning

confidence: 99%

“…This method is popular in image classification Tremblay et al (2018); Kortylewski et al (2018) and speech recognition Park et al (2019); Bird et al (2020) and it is oftern used in these areas. Similarly, in Feng et al (2020), DeepMM: Deep learning based map matching with data augmentation was proposed. The researchers proposed a statistic-based data augmentation and a generation-based data augmentation methods.…”

Section: Deep Learning In Transportation Engineeringmentioning

confidence: 99%

“…The length of the route is determined by the number of linked segments N. More complex routes are generated when N increases. In Feng et al (2020), authors first determine the origin and destination locations and then find the shortest path with adding spatial noise at some waypoints of the trajectory. However, we cannot guarantee that shortest path-based route generation method covers the whole routes in real situation.…”

Section: Road Network-based Synthetic Trajectory Generationmentioning

confidence: 99%

See 2 more Smart Citations

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Jin¹,

Choi²,

Yeo³

2021

Preprint

View full text Add to dashboard Cite

In many trajectory-based applications, it is necessary to map raw GPS trajectories onto road networks in digital maps, which is commonly referred to as a map-matching process. While most previous map-matching methods have focused on using rule-based algorithms to deal with the map-matching problems, in this paper, we consider the map-matching task from the data perspective, proposing a deep learning-based map-matching model. We build a Transformer-based map-matching model with a transfer learning approach. We generate synthetic trajectory data to pre-train the Transformer model and then fine-tune the model with a limited number of groundtruth data to minimize the model development cost and reduce the real-to-virtual gap. Three metrics (Average Hamming Distance, F-score, and BLEU) at two levels (point and segment level) are used to evaluate the model performance. The results indicate that the proposed model outperforms existing models. Furthermore, we use the attention weights of the Transformer to plot the map-matching process and find how the model matches the road segments correctly.

show abstract

Analytical Review of Map Matching Algorithms: Analyzing the Performance and Efficiency Using Road Dataset of the Indian Subcontinent

Singh

Goyal

et al. 2023

Arch Computat Methods Eng

View full text Add to dashboard Cite

DeepMM: Deep Learning Based Map Matching with Data Augmentation

Cited by 21 publications

References 27 publications

An explainable sentiment prediction model based on the portraits of users sharing representative opinions in social sensors

An explainable sentiment prediction model based on the portraits of users sharing representative opinions in social sensors

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Analytical Review of Map Matching Algorithms: Analyzing the Performance and Efficiency Using Road Dataset of the Indian Subcontinent

Contact Info

Product

Resources

About