Elastic Data Frame - Outlier Detection

| Menu | Next Post: Elastic Data Frame - Regression Analysis |

Unlike the Anomaly Detection models, this is a multi-variate analysis, it enables a better understanding of complex behaviors that are described by many features. For this analysis we have 3 models with different algorithms and learning types (Outlier, Regression and Classification) and in this post we'll talk about Outlier Detection.

Outlier detection identifies unusual data points in the dataset (Unsupervised ML).

When we talk about time series modeling and population anomaly detection, we look for outliers but basing it on how far the metric is from the normal model.  

With Outlier Detection we are looking at clusters of data and evaluating density and distance using multi-variate analysis. We are not interested in tracking evolution of this dataset over time like we do in population anomaly detection and there are no buckets.

Evaluation of the Outlier detection

Outliers may denote errors or unusual behavior. In the Elastic Stack, we use an ensemble of four different distance and density based outlier detection methods, based on this approach, a metric is computed called local outlier factor for each data point. The higher the local outlier factor, the more outlying is the data point.

| Menu | Next Post: Elastic Data Frame - Regression Analysis |

This post is part of a series that covers Artificial Intelligence with a focus on Elastic's (Creators of Elasticsearch) Machine Learning solution, aiming to introduce and exemplify the possibilities and options available, in addition to addressing the context and usability.