EXTRACT: Strong Examples from Weakly-Labeled Sensor Data
Open accessAbstract
Thanks to the rise of wearable and connected devices, sensor-generated time series comprise a large and growing fraction of the world's data. Unfortunately, extracting value from this data can be challenging, since sensors report low-level signals (e.g., acceleration), not the high-level events that are typically of interest (e.g., gestures). We introduce a technique to bridge this gap by automatically extracting examples of real-world events in low-level data, given only a rough estimate of when these events have taken place. By identifying sets of features that repeat in the same temporal arrangement, we isolate examples of such diverse events as human actions, power consumption patterns, and spoken words with up to 96% precision and recall. Our method is fast enough to run in real time and assumes only minimal knowledge of which variables are relevant or the lengths of events. Our evaluation uses numerous publicly available datasets and over 1 million samples of manually labeled sensor data.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Given rapid growth in adoption of Internet-enabled wearable devices, sensor time-series comprise a considerable amount of user-generated data.
“Given the rapid growth in adoption of Internet-enabled wearable devices, sensor time-series comprise a considerable amount of user-generated data [137].”
From Human-centred artificial intelligence for mobile health sensing: challenges and opportunities (2023) · cites this paper as [137]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Ye, L., & Keogh, E. (2009) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/1557019.1557122 · Cited by 992 · see citing works
Lin, J. (2002) · Cited by 446 · see citing works
Instructing people for training gestural interactive systems
Fothergill, S., Mentis, H. M., Kohli, P., & Nowozin, S. (2012) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/2207676.2208303 · Cited by 369 · see citing works
AMPds: A public dataset for load disaggregation and eco-feedback research
Makonin, S., Popowich, F. P., Bartram, L., Gill, B., & Bajić, I. V. (2013) · IEEE Electrical Power and Energy Conference (EPEC) · doi:10.1109/epec.2013.6802949 · Cited by 366 · see citing works
Clustering Time Series Using Unsupervised-Shapelets
Zakaria, J., Mueen, A., & Keogh, E. (2012) · Proceedings · doi:10.1109/icdm.2012.26 · Cited by 232 · see citing works
Detecting time series motifs under uniform scaling
Yankov, D., Keogh, E., Medina, J., Chiu, B., & Zordan, V. B. (2007) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/1281192.1281282 · Cited by 169 · see citing works
Flying Insect Classification with Inexpensive Sensors
Chen, Y., Why, A., Batista, G., Mafra‐Neto, A., & Keogh, E. (2014) · Journal of Insect Behavior · doi:10.1007/s10905-014-9454-4 · Cited by 152 · see citing works
Discovering Characteristic Actions from On-Body Sensor Data
Minnen, D., Starner, T., Essa, I., & Isbell, C. (2006) · 2006 10th IEEE International Symposium on Wearable Computers · doi:10.1109/iswc.2006.286337 · Cited by 132 · see citing works
Time Series Classification under More Realistic Assumptions
Hu, B., Chen, Y., & Keogh, E. (2013) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611972832.64 · Cited by 122 · see citing works
Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data
Rakthanmanon, T., Keogh, E., Lonardi, S., & Evans, S. (2011) · Proceedings · doi:10.1109/icdm.2011.146 · Cited by 119 · see citing works
Cheng, H.-T., Sun, F.-T., Griss, M., Davis, P., Li, J., & You, D. (2013) · International Conference on Mobile Systems, Applications, and Services · doi:10.1145/2462456.2464438 · Cited by 117 · see citing works
Enumeration of Time Series Motifs of All Lengths
Mueen, A. (2013) · Proceedings · doi:10.1109/icdm.2013.27 · Cited by 83 · see citing works
Discovering multivariate motifs using subsequence density estimation and greedy mixture learning
Minnen, D. C., Isbell, C. L., Essa, I. A., & Starner, T. E. (2007) · National Conference on Artificial Intelligence · Cited by 82 · see citing works
Minnen, D. C., Isbell, C. L., Essa, I. A., & Starner, T. E. (2007) · Proceedings · doi:10.1109/icdm.2007.52 · Cited by 81 · see citing works
Unsupervised Music Structure Annotation by Time Series Structure Features and Segment Similarity
Serrà, J., Müller, M., Grosche, P., & Arcos, J. L. (2014) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2014.2310701 · Cited by 69 · see citing works
Rare time series motif discovery from unbounded streams
Begum, N., & Keogh, E. (2014) · Proceedings of the VLDB Endowment · doi:10.14778/2735471.2735476 · Cited by 62 · see citing works
Improving activity discovery with automatic neighborhood estimation
Minnen, D. C., Starner, T. E., Essa, I. A., & Isbell, C. L. (2007) · Cited by 56 · see citing works
Scalable Clustering of Time Series with U-Shapelets
Ulanova, L., Begum, N., & Keogh, E. (2015) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611974010.101 · Cited by 55 · see citing works
Pattern discovery in data streams under the time warping distance
Toyoda, M., Sakurai, Y., & Ishikawa, Y. (2012) · The VLDB Journal · doi:10.1007/s00778-012-0289-3 · Cited by 47 · see citing works
Efficient Proper Length Time Series Motif Discovery
Yingchareonthawornchai, S., Sivaraks, H., Rakthanmanon, T., & Ratanamahatana, C. A. (2013) · Proceedings · doi:10.1109/icdm.2013.111 · Cited by 32 · see citing works
Dot plots for time series analysis
Yankov, D., Keogh, E., & Lonardi, S. (2005) · Proceedings - International Conference on Tools with Artificial Intelligence, TAI · doi:10.1109/ictai.2005.60 · Cited by 27 · see citing works
Parameter-free motif discovery for time series data
Nunthanid, P., Niennattrakul, V., & Ratanamahatana, C. A. (2012) · International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON) · doi:10.1109/ecticon.2012.6254126 · Cited by 22 · see citing works
G-SteX: Greedy Stem Extension for Free-Length Constrained Motif Discovery
Mohammad, Y. F. O., Ohmoto, Y., & Nishida, T. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31087-4_44 · Cited by 9 · see citing works
Multi-dimensional signal search with applications in remote medical monitoring
Moazeni, M., Mortazavi, B. J., & Sarrafzadeh, M. (2013) · IEEE International Conference on Body Sensor Networks (BSN) · doi:10.1109/bsn.2013.6575494 · Cited by 3 · see citing works
Flagged first, then by citations
1 of the 25 references couldn’t be matched to a known paper.
Research works printed but not linked: 2
- T. Oliphant, "Numba python bytecode to LLVM translator," in Proceed- ings of the Python for Scientific Computing Conference (SciPy), 2012.
- B. McFee, C. Raffel, D. Liang, D. P. Ellis, M. McVicar, E. Battenberg, and O. Nieto, "librosa: Audio and Music Signal Analysis in Python," in Proceedings of the 14th Python in Science Conference, 2015.
Cited by
Sorted by citations
Blalock, D. W., Madden, S. R., & Guttag, J. V. (2018) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3264903 · Cited by 146 · see citing works
Self-citationSelf-supervised transfer learning of physiological representations from free-living wearable data
Spathis, D., Perez-Pozuelo, I., Brage, S., Wareham, N. J., & Mascolo, C. (2021) · Proceedings of the Conference on Health, Inference, and Learning · doi:10.1145/3450439.3451863 · Cited by 40 · see citing works
Human-centred artificial intelligence for mobile health sensing: challenges and opportunities
Dang, T., Spathis, D., Ghosh, A., & Mascolo, C. (2023) · Royal Society Open Science · doi:10.1098/rsos.230806 · Cited by 30 · see citing works
Variable-Length Subsequence Clustering in Time Series
Duan, J., & Guo, L. (2020) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.2986965 · Cited by 13 · see citing works
Learning Generalizable Physiological Representations from Large-scale Wearable DataPreprint
Spathis, D., Perez-Pozuelo, I., Brage, S., Wareham, N. J., & Mascolo, C. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.04601 · Cited by 3 · see citing works
Adaptive Extraction of Variable-Length Subsequence Patterns in Noisy Time Series
Zhang, K., Duan, J., Yang, T., Wang, H., & Lv, C. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-8295-9_30
Adaptive variable-length subsequence pattern extraction in time series
Zhang, K., Duan, J., Yang, T., Guo, L., & Lv, C. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06752-3
- Authors
- Blalock, D. W., & Guttag, J. V.
- Affiliations
- Massachusetts Institute of Technology (US)
- ORCID
- Guttag, John V.
- Venue
- 2016 IEEE 16th International Conference on Data Mining (ICDM)
- Publisher
- IEEE
- Published
- 2016-12
- Pages
- 799-804
- ISSN
- 1550-4786 (ISSN-L)
- OpenAlex
- W2526992896
- Cited by
- 7
- References
- 25
- Citations by year
- 2025: 2 · 2023: 1 · 2021: 1 · 2020: 1 · 2018: 1 · 2016: 1
- FWCI
- 0.59
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- event extraction, wearable sensors, time series, gesture recognition, human action recognition, electricity consumption patterns, spoken word recognition, real-time processing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
