Projects

In the telecom industry, customers are able to choose from multiple service providers and actively switch from one operator to another. In this highly competitive market, the telecommunications industry experiences an average of 15-25% annual churn rate. Given the fact that it costs 5-10 times more to acquire a new customer than to retain an existing one, customer retention has now become even more important than customer acquisition.

For many incumbent operators, retaining high profitable customers is the number one business goal.

To reduce customer churn, telecom companies need to predict which customers are at high risk of churn.

In this project, you will analyse customer-level data of a leading telecom firm, build predictive models to identify customers at high risk of churn and identify the main indicators of churn.

Click here to read more!

Since the COVID19 world is transforming rapidly into a virtual and mobility-driven world, in parallel, software and intent-based network management have also become increasingly popular. There are two network planes in distributed networks: a "data plane" that processes and delivers packets with a local forwarding state and a "control plane" that computes the state of the routers and determines how and where the packets should be forwarded. The core traditional network infrastructure is made up of hardware in which the data and control planes are integrated within routers and switches, which has made maintaining the physical equipment as well as managing the data and control plane functionalities difficult. Also, there are other challenges, like scaling, securing and automating, which led to the development of the Software-Defined Network a decade ago. The Software-Defined Network segregates the control plane from the data plane in the switches and routers, thus enabling the management and orchestration of all the devices from the centralized SDN controller. Although SDN is rapidly evolving and shaping the next generation of heterogeneous network scenarios like datacenters, private enterprises, service providers, education campuses and home networks, they are vulnerable to new security threats which will allow attackers to take advantage of and exploit the network architecture to perform various intrusions. This dissertation aims to take the most recently developed attack specific SDN dataset and build a balancing technique to handle class imbalance, Feature Extraction methodology, build a classifier model using a deep learning algorithm to differentiate the attack traffic over the normal, and finally evaluate and compare the performance with the literature. Refer to the code here.