Argus provides a rich source of transactional network data that is suitable for many machine learning algorithms. Whether you are interested in credit card fraud approaches to cyber security, or pattern recognition to identify user network use behavior, Argus enables a platform that can support Bayesian deep learning, as well as supervised, unsupervised, semi-supervised learning, regression, instanced-based, regularization, decision tree, clustering, association rule learning, dimensional reduction, or ensemble analytics.
Python is a great platform for processing Argus data. Use Argus tools to generate general or specific network transactional data, collect, store, process, reduce, aggregate and extract the features needed for you ML analytics, and use Python and its libraries and functions, to analyze and visualize all that data.
These posts will describe in parts the steps needed to build an Argus ML workflow, reading data from Argus, inputing into a Python Panda’s dataframe, and then performing some basic exploratory data analysis.
These examples require the installation of specific Python packages which can be installed using the following commands:
pip install package-name
conda install package-name
To prepare Argus data for Python, simply generate a .csv file.
ra -r argus.file -c , > argus.csv
Here are some example projects / workflows that groups are working on.