Metadata-Version: 1.1
Name: categoricaloutlier
Version: 0.6.2
Summary: Trains on categorical and date time features of the data and predicts an anomaly score for new data
Home-page: https://github.com/akashbaj03/categoricaloutlier
Author: AKASH BAJPAI
Author-email: akash.baj03@gmail.com
License: MIT
Download-URL: https://github.com/akashbaj03/categoricaloutlier/archive/categoricaloutlier_v0.1.tar.gz
Description: Categorical Outlier is a tool to detect anomalous observations in categorical and DateTime features. Most of the techniques that we already have are focused on numeric features. There is no library available which can detect an outlier within categorical data. This package does that by building a profile using the past observations and gives an outlier score to a new observation on the basis of this profile. An example use of this package may be predicting unusual driving behavior. A driver who drives the same route(s) to reach maybe office will show an anomalous behavior if he takes an altogether different route on a particular day. He will get a high outlier score for this behavior. On the contrary, an uber driver drives to a new location every time and hence, a new destination will not be an anomalous behavior and hence will get a low score. The package also takes a combination of categorical features as input.
Keywords: categorical,outlier,anomaly,unsupervised,datetime,frequency,probability
Platform: UNKNOWN
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Topic :: Software Development :: Build Tools
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.4
Classifier: Programming Language :: Python :: 3.5
Classifier: Programming Language :: Python :: 3.6
