Telecom Network Anomaly Detection

Network security is becoming increasingly important as large number of applications are running on the top of it. There are number of network parameters that are commonly used to specify the Telecom network. Careful analysis is required to be done in order to understand the vulnerability of the network. Therefore anomaly detection is so important in the network. The data science techniques are used to study the network parameters and then classification algorithms are applied to classify the network issues. The web based application is also developed using Streamlit so that the user can interact with the application by changing the network parameters thereby helping the user to understand whether the network is ‘OK’ or ‘Not OK’.

Salient Features:
– Detailed study of network parameters and relationships
– Feature engineering
– Training and evaluation of ML models for classification
– Feature importance
– Simple web application to demonstrate the anomaly in the network

Tech Stacks:
Numpy, Pandas, Scikit Learn, Matplotlib, Seaborn, Python, Streamlit

Industry:
Telecom