EXTRACT: Strong Examples from Weakly-Labeled Sensor Data
Authors:
- Davis W. BlalockDBDavis W. Blalock
- John V. GuttagJGJohn V. Guttag
- 1 December 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 799-804
Abstract
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.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- EXTRACT: Strong Examples from Weakly-Labeled Sensor Data2016
- Rare time series motif discovery from unbounded streamsProceedings of the VLDB Endowment, 2014
- Enumeration of Time Series Motifs of All Lengths2013
- Efficient Proper Length Time Series Motif Discovery2013
- AMPds: A public dataset for load disaggregation and eco-feedback research2013
- NuActiv2013
- Parameter-free motif discovery for time series data2012
- Time Series Epenthesis: Clustering Time Series Streams Requires Ignoring Some Data2011
- Time series shapelets2009
- Detecting time series motifs under uniform scaling2007