Implementing Azure Machine Learning Platform
Executive Summary
This client is an upcoming, local consumer credit reporting agency that handles over a million individual consumers. Initially, their analysts were individually and manually checking each alert. They realized the inefficacy of this method and reached out to us for a solution to automating the checking of each individual alert. We were able to cut down their time 83% and enable greater efficiency in monitoring alerts.
The Business Need
The client was searching for a solution to increase the efficiency of their analysts to manage and triage a flood of alerts from over 400 hard-coded rules in Splunk in order to separate false positives from true alerts. The client wanted to automate the process of checking each alert against other suspicious activities such as unusual increases in files being transferred, spikes in network traffic, and attempts to reach other known bad URLs. They were looking for a partner to help build a workflow that would speed up the process and decrease analyst work hours.
The Solution
We discussed with our client to better understand their workflow needs. Using Microsoft Azure's machine learning platform, we were able to produce an automated workflow which mimicked all the steps an analyst would have to take to assess each alert.
The Result
Our automated workflow we created was able to make the clients process of differentiating alerts much more efficient and productive. This resulted in cutting down the processing time of each alert from 30 mins to only 5 mins.