
Communications, Media & Telecom - Retail Commission Diagnostics Anomaly Detection
Communications, Media & Telecom
A telecommunications company faced anomalies in reported sales figures, requiring manual checks. synvert built an Oracle-based architecture, and trained a Machine Learning model to detect and visualize these anomalies, saving time and improving revenue accuracy.
Initial situation
The customer was experiencing unusual behavior where the reported sales figures were occasionally anomalous.
For this reason, it was necessary to manually check how much sales were made and whether there were any anomalies in the original figures.
Architecture
The architecture is built on Oracle using technologies such as Pyspark, Sklearn, Pandas, and Tableau.
Generated benefits
It saved time by eliminating manual error checking and allowed for a more accurate knowledge of revenue.
Services accomplished by synvert
synvert trained a Machine Learning model to detect such anomalies and visualize them to the users in a dashboard.