1
$\begingroup$

I am looking for algorithms on Anomaly detection for time series data. It is uni-variate analysis, considering single parameter (inlet pressure) of air compressor sensor data. The objective is to detect if there is any significant deviation in parameter value in real time.

If any one has worked on similar projects, please share your thoughts.

Improve this question
$\endgroup$
3

1 Answer 1

Reset to default
1
$\begingroup$

Welcome to DS forum. There are many methods available for anomaly detection of uni-variate, low dimension and high dimension data sets.

Supervised algorithms can be used if the past anomalies are recorded with flag variables and there is significant examples of flagged anomalies.

In unsupervised algorithms the key aspect of anomaly is which subset of the larger data set is considered when looking for anomolies i.e., global vs local and point vs contextual

Anomaly Types

Here are some useful links to get you started, anomaly detection in machine generated data and introductory overview of time-series based anomaly detection algorithms.

Hope this helps.

Improve this answer
$\endgroup$
4
  • $\begingroup$ Manohar Swamynathan and Laurent Duval. Thanks for your response. The analysis is unsupervised in nature and considering a single variable (inlet pressure). The objective is to develop a model which can detect if there is any significant deviation in the parameter value in real time. $\endgroup$
    – Sunil M
    Apr 27, 2016 at 9:17
  • $\begingroup$ @ Sunil for your use case, the uni-variate methods discussed in the paper anomaly detection in machine generated data has been implemented in python and the same can be found in here $\endgroup$
    – Manohar Swamynathan
    Apr 27, 2016 at 10:36
  • $\begingroup$ @ Manohar, thanks for the document. Ya, I am going through the material. Can you let me know if there is any document where the method is implemented in R ? $\endgroup$
    – Sunil M
    Apr 27, 2016 at 11:12
  • $\begingroup$ I'm not sure of implementation of this paper content in R, however you can look at AnomalyDetection R package and also there is a new package by Rob J Hyndman Anomalous time series package. $\endgroup$
    – Manohar Swamynathan
    Apr 27, 2016 at 14:44

Not the answer you're looking for? Browse other questions tagged or ask your own question.