0
$\begingroup$

I want to build a system that will detect Anomaly for categorical data. I have a timeseries data like this enter image description here For metric data these anomalies are calculated

  1. Outliers detection
  2. Trend
  3. Pattern
  4. Change

I have no idea what & how do I need to calculate for finding anomalies in categorical data. Result for the table may look like something "Everyday at 10am there is success, but today it is failure" related to these. Can someone give me some lead, so that I will get started on this.

Improve this question
$\endgroup$
9
  • $\begingroup$ Trend, change. .. : are you looking at time series data? Everyday... : maybe even time series with repetitions? $\endgroup$
    – Ute
    Commented Jul 14, 2023 at 6:42
  • $\begingroup$ Yes Ute, It is a timeseries data $\endgroup$
    – Raj
    Commented Jul 14, 2023 at 6:42
  • $\begingroup$ This is important information. Although one can guess it perpaps from the table you show, it would help if you give this information literally in your question :-) $\endgroup$
    – Ute
    Commented Jul 14, 2023 at 6:45
  • 1
    $\begingroup$ Updated the question. Thanks for pointing that $\endgroup$
    – Raj
    Commented Jul 14, 2023 at 6:48
  • 1
    $\begingroup$ Great! If you are at the very beginning and need inspiration, your magic Google key word is "categorical time series". You can add "anomaly detection" to it $\endgroup$
    – Ute
    Commented Jul 14, 2023 at 7:01

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.