EXTRACT: Strong Examples from Weakly-Labeled Sensor Data

TL;DR

EXTRACT is a technique that automatically identifies high-level events from low-level sensor data using minimal prior information, achieving high precision and recall in real-time applications across diverse datasets.

Contribution

The paper introduces a novel method for extracting real-world events from weakly-labeled sensor data without extensive prior knowledge, enabling real-time analysis.

Findings

Achieves up to 96% precision and recall in event extraction.

Works in real time with minimal knowledge of relevant variables.

Validated on over 1 million labeled sensor samples.

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.