Using GPS to learn significant locations and predict movement across multiple users

By Daniel Ashbrook, Thad Starner. Personal and Ubiquitous Computing, 2003

1,236 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Transportation, Signal Processing

View Paper: https://www.semanticscholar.org/paper/2578219742857f4af310b003c03f607ed01d79f8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Robert K. Harle
2. Yu Zheng
3. Dieter Fox
4. Nicholas Donald Lane
5. Andrew T. Campbell
6. Deborah Estrin
7. Xing Xie
8. Oliver J. Woodman

In Transportation

1. Yu Zheng
2. Xing Xie
3. Wei-Ying Ma
4. Cecilia Mascolo
5. Mirco Musolesi
6. Yukun Chen
7. Lizhu Zhang
8. Quannan Li

In Signal Processing

1. Yu Zheng
2. Xing Xie
3. Wei-Ying Ma
4. Robert K. Harle
5. Deborah Estrin
6. Yukun Chen
7. Lizhu Zhang
8. Quannan Li

Citations in the corpus

• Understanding mobility based on GPS data (2008, background): “These works include extracting significant places of an individual [3][7][8], predicting a person’s movement [6][10] and modeling a user’s transportation routine [8][10].”
• A clustering-based approach for discovering interesting places in trajectories (2008, methodology): “Ashbrook and Starner [3] use the well-know clustering algorithm K-means to find important places in trajectories.”
• Identifying intercity freight trip ends of heavy trucks from GPS data (2021, methodology): “Recently, a variety of spatial clustering algorithms have been proposed, such as an improved K-means algorithm [35], DJCluster [27], CB-SMoT [24], T-DBSCAN [29], DBSCAN-TE [36], and TAD [26].”
• Learning travel recommendations from user-generated GPS traces (2011, neutral): “These works include detecting signi.cant locations of a user [Ashbrook and Starner 2003; Hariharan and Toyama 2004], predicting the user s movement among these locations [Liao et al 2005], and recognizing user­speci.c activities at each location [Liao et al. 2004; Patterson et al.”
• Predicting User-Cell Association in Cellular Networks from Tracked Data (2009, background): “Ashbrook and Starner showed how locations of significance could be automatically learned from GPS data at multiple scales [3].”
• Recommender Systems (2016, neutral): “One of the earliest works [54] proposes methods to use the data from GPS-enabled mobile phones to predict user movement across various locations.”
• Mining interesting locations and travel sequences from GPS trajectories (2009, background): “These works include detecting significant locations of a user [5][9], predicting the user’s movement among these locations [ 5 ][13], and recognizing user-specific activities at each location [15].”
• Practical Metropolitan-Scale Positioning for GSM Phones (2006, background): “Instead, they provide the ability to recognize previously visited destinations by using metrics like GPS dropout [2], cell handoff patterns [8], or radio signatures [7] to match previously learned places.”
• Extracting places from traces of locations (2004, methodology): “Ashbrook and Starner [1] did not have a clear knee in their curve and chose a value of 600 seconds somewhat arbitrarily.”