Location-Based Activity Recognition with Hierarchical Dirichlet Process
Abstract
We consider the problem of analyzing people's mobility and movement patterns from their location history, gathered by mobile devices. Human mobility traces can be extremely complex and unpredictable, by nature, which makes it hard to construct accurate models of mobility behavior. In this work, we present a novel high-level strategy for mobility data analysis based on Hierarchical Dirichlet process, which is a powerful probabilistic model for clustering grouped data. We evaluate our unsupervised approach on two real-world datasets. PDF
Cite
Text
Ghourchian. "Location-Based Activity Recognition with Hierarchical Dirichlet Process." International Joint Conference on Artificial Intelligence, 2016.Markdown
[Ghourchian. "Location-Based Activity Recognition with Hierarchical Dirichlet Process." International Joint Conference on Artificial Intelligence, 2016.](https://mlanthology.org/ijcai/2016/ghourchian2016ijcai-location/)BibTeX
@inproceedings{ghourchian2016ijcai-location,
title = {{Location-Based Activity Recognition with Hierarchical Dirichlet Process}},
author = {Ghourchian, Negar},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2016},
pages = {3990-3991},
url = {https://mlanthology.org/ijcai/2016/ghourchian2016ijcai-location/}
}