Abstract
Recent hurricane events have caused unprecedented amounts of damage on critical infrastructure systems and have severely threatened our public safety and economic health. The most observable (and severe) impact of these hurricanes is the loss of electric power in many regions, which causes breakdowns in essential public services. Understanding power outages and how they evolve during a hurricane provides insights on how to reduce outages in the future, and how to improve the robustness of the underlying critical infrastructure systems. In this article, we propose a novel scalable segmentation with explanations framework to help experts understand such datasets. Our method, CnR (Cut-n-Reveal), first finds a segmentation of the outage sequences based on the temporal variations of the power outage failure process so as to capture major pattern changes. This temporal segmentation procedure is capable of accounting for both the spatial and temporal correlations of the underlying power outage process. We then propose a novel explanation optimization formulation to find an intuitive explanation of the segmentation such that the explanation highlights the culprit time series of the change in each segment. Through extensive experiments, we show that our method consistently outperforms competitors in multiple real datasets with ground truth. We further study real county-level power outage data from several recent hurricanes (Matthew, Harvey, Irma) and show that CnR recovers important, non-trivial, and actionable patterns for domain experts, whereas baselines typically do not give meaningful results.
Supplemental Material
Available for Download
Supplemental movie, appendix, image and software files for, Cut-n-Reveal: Time Series Segmentations with Explanations
- Boston Herald. 2017. Hurricane Harvey. Retrieved June 16, 2020 from https://www.bostonherald.com/2017/08/25/hurricane-harvey-to-slam-texas-coast-hard/.Google Scholar
- Carnegie Mellon University. 2014. CMU Graphics Lab Motion Capture Database. Retrieved June 16, 2020 from http://mocap.cs.cmu.edu.Google Scholar
- Dekalb County Georgia. 2017. Georgia Power Working to Restore Service to 161,000 in DeKalb. Retrieved June 16, 2020 from https://www.dekalbcountyga.gov/news/georgia-power-working-restore-service-161000-dekalb.Google Scholar
- 2017. Major Storm System Impacting Holiday Travel Through Friday. Retrieved June 16, 2020 from https://www.weather.gov/crp/hurricane_harvey.Google Scholar
- NOAA’s National Weather Service. 2017. Severe Weather Event Review for Saturday August 26 2017. Retrieved June 16, 2020 from http://www.spc.noaa.gov/exper/archive/event.php?date=20170826.Google Scholar
- 2019. Appendix. Retrieved June 16, 2020 from https://bit.ly/2JVt8GP.Google Scholar
- GitHub. 2020. Code and Datasets. Retrieved June 16, 2020 from https://github.com/anikat1/cnr-tist/blob/master/appendix.pdf.Google Scholar
- M. Allen, S. Fernandez, O. Omitaomu, and K. Walker. 2014. Application of hybrid geo-spatially granular fragility curves to improve power outage predictions. Journal of Geography & Natural Disasters 4, 127 (2014), 2167--0587.Google Scholar
- Samaneh Aminikhanghahi and Diane J. Cook. 2017. A survey of methods for time series change point detection. Knowledge and Information Systems 51, 2 (2017), 339--367.Google ScholarDigital Library
- Francis Bach, Rodolphe Jenatton, Julien Mairal, and Guillaume Obozinski. 2011. Convex optimization with sparsity-inducing norms. Optimization for Machine Learning 5 (2011), 19--53.Google Scholar
- Oresti Banos, Mate Attila Toth, Miguel Damas, Hector Pomares, and Ignacio Rojas. 2014. Dealing with the effects of sensor displacement in wearable activity recognition. Sensors 14, 6 (2014), 9995--10023.Google ScholarCross Ref
- Alan M. Barker, Eva B. Freer, Olufemi A. Omitaomu, Steven J. Fernandez, Supriya Chinthavali, and Jeffrey B. Kodysh. 2013. Automating natural disaster impact analysis: An open resource to visually estimate a hurricane’s impact on the electric grid. In Proceedings of the IEEE Southeast Conference. IEEE, Los Alamitos, CA, 1--3.Google Scholar
- Richard H. Bartels and George W. Stewart. 1972. Solution of the matrix equation AX + XB = C [F4]. Communications of the ACM 15, 9 (1972), 820--826.Google ScholarDigital Library
- Durell Bouchard. 2006. Automated Time Series Segmentation for Human Motion Analysis. Center for Human Modeling and Simulation, University of Pennsylvania.Google Scholar
- Stephen Boyd, Neal Parikh, Eric Chu, Borja Peleato, and Jonathan Eckstein. 2011. Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundations and Trends® in Machine Learning 3, 1 (2011), 1--122.Google ScholarDigital Library
- Liang Chang and Zhigang Wu. 2011. Performance and reliability of electrical power grids under cascading failures. International Journal of Electrical Power & Energy Systems 33, 8 (2011), 1410--1419.Google ScholarCross Ref
- Liangzhe Chen, Sorour E. Amiri, and B. Aditya Prakash. 2018. Automatic Segmentation of Data Sequences. Association for the Advancement of Artificial Intelligence.Google Scholar
- Liangzhe Chen, Xinfeng Xu, Sangkeun Lee, Sisi Duan, Alfonso G. Tarditi, Supriya Chinthavali, and B. Aditya Prakash. 2017. HotSpots: Failure cascades on heterogeneous critical infrastructure networks. In Proceedings of the 2017 ACM Conference on Information and Knowledge Management (CIKM’17). ACM, New York, NY, 1599--1607.Google Scholar
- Eagle. 2012. Eagle-I. Retrieved June 16, 2020 from https://eagle-i.doe.gov/.Google Scholar
- Ehsan Elhamifar and René Vidal. 2009. Sparse subspace clustering. In Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’09). IEEE, Los Alamitos, CA, 2790--2797.Google ScholarCross Ref
- Rozhin Eskandarpour and Amin Khodaei. 2017. Machine learning based power grid outage prediction in response to extreme events. IEEE Transactions on Power Systems 32, 4 (2017), 3315--3316.Google ScholarCross Ref
- Zoubin Ghahramani. 2015. Probabilistic machine learning and artificial intelligence. Nature 521, 7553 (2015), 452.Google Scholar
- Shaghayegh Gharghabi, Yifei Ding, Chin-Chia Michael Yeh, Kaveh Kamgar, Liudmila Ulanova, and Eamonn Keogh. 2017. Matrix profile VIII: Domain agnostic online semantic segmentation at superhuman performance levels. In Proceedings of the 2017 IEEE International Conference on Data Mining (ICDM’17). IEEE, Los Alamitos, CA, 117--126.Google ScholarCross Ref
- Gene Golub, Stephen Nash, and Charles Van Loan. 1979. A Hessenberg-Schur method for the problem AX + XB = C. IEEE Transactions on Automatic Control 24, 6 (1979), 909--913.Google ScholarCross Ref
- David Hallac, Sagar Vare, Stephen Boyd, and Jure Leskovec. 2017. Toeplitz inverse covariance-based clustering of multivariate time series data. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’17). ACM, New York, NY, 215--223.Google ScholarDigital Library
- Paul Hines, Karthikeyan Balasubramaniam, and Eduardo Cotilla Sanchez. 2009. Cascading failures in power grids. IEEE Potentials 28, 5 (2009), 24--30.Google ScholarCross Ref
- Paul D. H. Hines, Ian Dobson, and Pooya Rezaei. 2017. Cascading power outages propagate locally in an influence graph that is not the actual grid topology. IEEE Transactions on Power Systems 32, 2 (2017), 958--967.Google Scholar
- Åke J. Holmgren. 2006. Using graph models to analyze the vulnerability of electric power networks. Risk Analysis 26, 4 (2006), 955--969.Google ScholarCross Ref
- Wei Hong, John Wright, Kun Huang, and Yi Ma. 2006. Multiscale hybrid linear models for lossy image representation. IEEE Transactions on Image Processing 15, 12 (2006), 3655--3671.Google ScholarDigital Library
- Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer 42, 8 (2009), 30--37.Google ScholarDigital Library
- Colin Lea, Austin Reiter, René Vidal, and Gregory D. Hager. 2016. Segmental spatiotemporal CNNs for fine-grained action segmentation. In Proceedings of the European Conference on Computer Vision (ECCV’16). 36--52.Google Scholar
- Lei Li, James McCann, Nancy S. Pollard, and Christos Faloutsos. 2009. DynaMMo: Mining and summarization of coevolving sequences with missing values. In Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’09). ACM, New York, NY, 507--516.Google ScholarDigital Library
- Junmin Liu, Yijun Chen, Jiangshe Zhang, and Zongben Xu. 2014. Enhancing low-rank subspace clustering by manifold regularization. IEEE Transactions on Image Processing 23, 9 (2014), 4022--4030.Google ScholarCross Ref
- Jun Liu and Jieping Ye. 2010. Efficient l1/lq norm regularization. arXiv:1009.4766.Google Scholar
- Yasuko Matsubara, Yasushi Sakurai, and Christos Faloutsos. 2014. AutoPlait: Automatic mining of co-evolving time sequences. In Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data (SIGMOD’14). ACM, New York, NY, 193--204.Google ScholarDigital Library
- Abdullah Mueen and Eamonn Keogh. 2010. Online discovery and maintenance of time series motifs. In Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’10). 1089--1098.Google ScholarDigital Library
- Nikhil Muralidhar, Chen Wang, Nathan Self, Marjan Momtazpour, Kiyoshi Nakayama, Ratnesh Sharma, and Naren Ramakrishnan. 2018. illiad: InteLLigent invariant and anomaly detection in cyber-physical systems. ACM Transactions on Intelligent Systems and Technology 9, 3 (2018), 35.Google ScholarDigital Library
- Xia Ning and George Karypis. 2011. SLIM: Sparse linear methods for top-n recommender systems. In Proceedings of the 2011 IEEE 11th International Conference on Data Mining (ICDM’11). IEEE, Los Alamitos, CA, 497--506.Google ScholarDigital Library
- Min Ouyang. 2014. Review on modeling and simulation of interdependent critical infrastructure systems. Reliability Engineering & System Safety 121 (2014), 43--60.Google ScholarCross Ref
- Lance Parsons, Ehtesham Haque, and Huan Liu. 2004. Subspace clustering for high dimensional data: A review. ACM SIGKDD Explorations Newsletter 6, 1 (2004), 90--105.Google ScholarDigital Library
- Forough Poursabzi-Sangdeh, Daniel G. Goldstein, Jake M. Hofman, Jennifer Wortman Vaughan, and Hanna Wallach. 2018. Manipulating and measuring model interpretability. arXiv:1802.07810.Google Scholar
- Naren Ramakrishnan, Satish Tadepalli, Layne T. Watson, Richard F. Helm, Marco Antoniotti, and Bud Mishra. 2010. Reverse engineering dynamic temporal models of biological processes and their relationships. Proceedings of the National Academy of Sciences 107, 28 (2010), 12511--12516.Google ScholarCross Ref
- Jaxk Reeves, Jien Chen, Xiaolan L. Wang, Robert Lund, and Qi Qi Lu. 2007. A review and comparison of changepoint detection techniques for climate data. Journal of Applied Meteorology and Climatology 46, 6 (2007), 900--915.Google ScholarCross Ref
- Andreas Reinhardt, Paul Baumann, Daniel Burgstahler, Matthias Hollick, Hristo Chonov, Marc Werner, and Ralf Steinmetz. 2012. On the accuracy of appliance identification based on distributed load metering data. In Proceedings of the 2012 Conference on Sustainable Internet and ICT for Sustainability (SustainIT’12). IEEE, Los Alamitos, CA, 1--9.Google Scholar
- Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. “Why should I trust you?”: Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’16). ACM, New York, NY, 1135--1144.Google ScholarDigital Library
- Guy Rosman, Mikhail Volkov, Dan Feldman, John W. Fisher III, and Daniela Rus. 2014. Coresets for k-segmentation of streaming data. In Proceedings of the 27th International Conference on Neural Information Processing Systems—Volume 1 (NIPS’14). 559--567.Google ScholarDigital Library
- Eric Ruggieri. 2013. A Bayesian approach to detecting change points in climatic records. International Journal of Climatology 33, 2 (2013), 520--528.Google ScholarCross Ref
- Allou Samé and Gérard Govaert. 2012. Online time series segmentation using temporal mixture models and Bayesian model selection. In Proceedings of the 2012 11th International Conference on Machine Learning and Applications (ICMLA’12). 602--605.Google ScholarDigital Library
- Rishu Saxena, Layne T. Watson, Randolph H. Wynne, Evan B. Brooks, Valerie A. Thomas, Yang Zhiqiang, and Robert E. Kennedy. 2018. Towards a polyalgorithm for land use change detection. ISPRS Journal of Photogrammetry and Remote Sensing 144 (2018), 217--234.Google ScholarCross Ref
- Jianbo Shi and Jitendra Malik. 2000. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 22, 8 (2000), 888--905.Google ScholarDigital Library
- Yuliya Tarabalka, Guillaume Charpiat, Ludovic Brucker, and Bjoern H. Menze. 2014. Spatio-temporal video segmentation with shape growth or shrinkage constraint. IEEE Transactions on Image Processing 23, 9 (2014), 3829--3840.Google ScholarCross Ref
- Stephen Tierney, Junbin Gao, and Yi Guo. 2014. Subspace clustering for sequential data. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’14). 1019--1026.Google ScholarDigital Library
- René Vidal. 2011. Subspace clustering. IEEE Signal Processing Magazine 28, 2 (2011), 52--68.Google ScholarCross Ref
- Xian Wu, Yuxiao Dong, Chao Huang, Jian Xu, Dong Wang, and Nitesh V. Chawla. 2017. UAPD: Predicting urban anomalies from spatial-temporal data. In Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases. 622--638.Google Scholar
- Allen Y. Yang, John Wright, Yi Ma, and S. Shankar Sastry. 2008. Unsupervised segmentation of natural images via lossy data compression. Computer Vision and Image Understanding 110, 2 (2008), 212--225.Google ScholarDigital Library
- Huaxiu Yao, Xianfeng Tang, Hua Wei, Guanjie Zheng, and Zhenhui Li. 2019. Revisiting spatial-temporal similarity: A deep learning framework for traffic prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 33. 5668--5675.Google ScholarDigital Library
- Yinyu Ye and Edison Tse. 1989. An extension of Karmarkar’s projective algorithm for convex quadratic programming. Mathematical Programming 44, 1--3 (1989), 157--179.Google ScholarDigital Library
- Xin Zhao and Pao-Shin Chu. 2006. Bayesian multiple changepoint analysis of hurricane activity in the eastern North Pacific: A Markov chain Monte Carlo approach. Journal of Climate 19, 4 (2006), 564--578.Google ScholarCross Ref
Index Terms
- Cut-n-Reveal: Time Series Segmentations with Explanations
Recommendations
Learning Social Meta-knowledge for Nowcasting Human Mobility in Disaster
WWW '23: Proceedings of the ACM Web Conference 2023Human mobility nowcasting is a fundamental research problem for intelligent transportation planning, disaster responses and management, etc. In particular, human mobility under big disasters such as hurricanes and pandemics deviates from its daily ...
Inferring Causal Interactions in Financial Markets Using Conditional Granger Causality Based on Quantile Regression
AbstractGranger causality analysis emerges as a typical method for inferring causal interactions in economics variables. Yet the traditional pairwise approach to Granger causality analysis may not clearly distinguish between direct causal influences from ...
Motion estimation and segmentation method based on integration of spatial and temporal probability models
ASID'09: Proceedings of the 3rd international conference on Anti-Counterfeiting, security, and identification in communicationA novel video motion object automatic segmentation algorithm based on a Bayesian framework is studied in this paper. A fast estimation procedure for the posterior marginals is added to the MAP algorithm. The field is initialized as the temporal ...
Comments