Bridging the Gap between Anomaly Detection and Action with Visual Analytics

Abstract

Quals talk:

Anomaly detection is widely used in business applications due to the ever-present possibility of costly operational failures. However, resolving failures requires the communication of actionable insights to decision makers, going beyond the raw output of existing anomaly detection algorithms. This gap in technology forces data scientists to rely on tedious manual approaches to constructing actionable generalizations to make sense of the found anomalies and communicate the results.

In this talk I will discuss how visual analytics systems can be paired with anomaly detection algorithms to bridge the gap between the detection of anomalies and the communication of actionable insights. I will begin by presenting a proposed path from anomalies to action derived from interviews with data scientists. Next, I will discuss PIXAL, a visualization analytics system developed to facilitate the generation and exploration of anomaly explanations. Finally, I will discuss future research addressing the validation and communication of anomalies.

Please join the meeting in Sococo VH 102, or Zoom.

Join Zoom Meeting: https://tufts.zoom.us/j/98610939077

PASSWORD: See colloquia email

Dial by your location: +1 646 558 8656 US (New York)

Meeting ID: 986 1093 9077

PASSCODE: See colloquia email