A Unified Framework for Robust and Efficient Hotspot Detection in Smart Cities

A Unified Framework for Robust and Efficient Hotspot Detection in Smart Cities
复制标题

智慧城市中稳健高效的热点检测的统一框架

DOI:
10.1145/3379562
复制
发表时间:
2020
期刊:
ACM/IMS Transactions on Data Science
影响因子:
--
通讯作者:
Shekhar, Shashi
Shekhar, Shashi
中科院分区:
--
文献类型:
--
作者:
Xie, Yiqun;Shekhar, Shashi

文献摘要

参考文献

相似文献

给定N个地理定位点实例(例如,犯罪或疾病病例),我们的目标是检测子区域(即,热点),其具有比其它实例更高的生成这样的实例的概率密度。热点检测已被广泛用于各种重要的城市应用,包括公共安全、公共卫生、城市规划和公平等。这个问题是具有挑战性的,因为它的社会应用通常对假阳性的容忍度很低,并且需要计算密集型的显著性测试。在相关的工作中,空间扫描统计量引入了一个基于似然比的框架,用于热点评估和显著性测试。然而,它没有考虑空间非确定性的影响,导致许多漏检测。我们之前的工作引入了一个基于非确定性标准化的扫描统计来缓解这个问题。然而,其对假阳性的鲁棒性并没有得到稳定的控制。为了解决这些限制,我们提出了一个统一的框架,可以提高结果的完整性,而不会产生更多的误报。我们还提出了一个约简算法,以提高计算效率。实验结果表明,该统一框架在不增加误报率的情况下,能显著提高热点检测的查全率,而约简算法能显著减少执行时间。
GivenNgeo-located point instances (e.g., crime or disease cases) in a spatial domain, we aim to detect sub-regions (i.e., hotspots) that have a higher probability density of generating such instances than the others. Hotspot detection has been widely used in a variety of important urban applications, including public safety, public health, urban planning, and equity, among others. The problem is challenging because its societal applications often have low tolerance for false positives and require significance testing that is computationally intensive. In related work, the spatial scan statistic introduced a likelihood ratio--based framework for hotspot evaluation and significance testing. However, it fails to consider the effect of spatial non-determinism, causing many missing detections. Our previous work introduced a non-deterministic normalization--based scan statistic to mitigate this issue. However, its robustness against false positives is not stably controlled. To address these limitations, we propose a unified framework that can improve the completeness of results without incurring more false positives. We also propose a reduction algorithm to improve the computational efficiency. Experiment results confirm that the unified framework can greatly improve the recall of hotspot detection without increasing the number of false positives, and the reduction algorithm can greatly reduce execution time.
基于非确定性归一化的扫描统计(NN-scan)实现稳健的热点检测:结果总结
DOI: 10.1137/1.9781611975673.10
发表时间: 2019
期刊: Proceedings of the 2019 SIAM International Conference on Data Mining
影响因子: --
作者:
Xie, Yiqun;Shekhar, Shashi.
通讯作者: Shekhar, Shashi.
DOI: 10.1111/j.1467-9671.2010.01194.x
发表时间: 2010-06-01
影响因子: 2.4
作者:
Nakaya, Tomoki;Yano, Keiji
通讯作者: Yano, Keiji
椭圆空间扫描统计及其在美国东北部乳腺癌死亡率数据中的应用
DOI: --
发表时间: 2003
期刊: Journal of urban health
影响因子: --
作者:
M. Kulldorff;Lan Huang;L. Pickle
通讯作者: L. Pickle
DOI: 10.1109/isc2.2018.8656800
发表时间: 2018-09
期刊: 2018 IEEE International Smart Cities Conference (ISC2)
影响因子: --
作者:
Yiqun Xie;Jayant Gupta;Yan Li;S. Shekhar
通讯作者: Yiqun Xie;Jayant Gupta;Yan Li;S. Shekhar
DOI: 10.1109/tbdata.2016.2631518
发表时间: 2017-06
影响因子: 7.2
作者:
Xun Tang;E. Eftelioglu;Dev Oliver;S. Shekhar
通讯作者: Xun Tang;E. Eftelioglu;Dev Oliver;S. Shekhar