Yukihiro, Hayley; 2021-09-03 Friday

Vision

Purpose of task: to introduce/familiarize with industry terms to describe the "health" of units

Explore the following inquiries:

Inquiry	Thought process
What other YouTube links can you point the team toward to better understand data?	drop links in the Toby extension, and drop a 1-3 sentence summary or bullet list of highlights
What language/dictionary can we use/create to characterize different units?	"Dead", "Beating", "Intermittent"? Are these words sufficiently descriptive? What other words might we consider?

We will whiteboard this subject on Friday to discuss previous experience with classifying the "health" of units, along with research insight gained from this task.

Summary of Research 9/3:

• Anomaly detection (aka outlier analysis) is a step in data mining that identifies data points, events, and/or observations that deviate from a dataset’s **normal behavior **
• key performance indicators (KPIs) that evaluate the success of the organization
  • KPIs can determine the expected or normal behavior of the data
• The absence of change can be an anomaly if it breaks a pattern that is normal for the data from that particular metric.
  • what are the insights for our data
  • we determine the normal metric
  • the metric can be either frequency (black Friday) or a specific value
• Anomalies aren’t categorically good or bad, they’re just deviations from the expected value for a metric at a given point in time.
• Time series data contains information necessary for making educated guesses about what can be reasonably **expected **in the future
  • Like blueprinting in cybersecurity, we need to become familiar with the data to determine what is normal → then we determine what is healthy data
  • Time series data anomaly detection must first create a baseline for normal behavior in primary KPIs
• Understanding the types of outliers that an anomaly detection system can identify is essential to getting the most value from generated insights.
• By discovering **normal tendency **or dealing out of data usually we find information but infrequent occurrence or data item sometimes may provide details which is very useful to us.
• Anomaly detection is a technique used to identify unusual patterns that do not conform to expected behavior, called outliers.