Timezone: »

 
Poster
Precision and Recall for Time Series
Nesime Tatbul · Tae Jun Lee · Stan Zdonik · Mejbah Alam · Justin Gottschlich

Tue Dec 04 02:00 PM -- 04:00 PM (PST) @ Room 517 AB #116
in Tue Poster Session B »

Classical anomaly detection is principally concerned with point-based anomalies, those anomalies that occur at a single point in time. Yet, many real-world anomalies are range-based, meaning they occur over a period of time. Motivated by this observation, we present a new mathematical model to evaluate the accuracy of time series classification algorithms. Our model expands the well-known Precision and Recall metrics to measure ranges, while simultaneously enabling customization support for domain-specific preferences.

Keywords: Time Series Analysis · Predictive Models · Classification · Benchmarks · Activity and Event Recognition

Author Information

Nesime Tatbul (Intel Labs and MIT)
Tae Jun Lee (Microsoft)
Stan Zdonik (Brown University)
Mejbah Alam (Intel Labs)
Justin Gottschlich (Intel Labs)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors