Home

I am a Data Scientist at Telstra. I am part of the advanced analytics and modelling team within the CMO. I have developed predictive models for targeting customers for cross-sell and retention campaigns. I have also conducted a number of deep-dive analyses that have produced valuable insights for the business. One of my areas of focus is customer advocacy (NPS), specifically to conduct analyses with the goal of gaining deeper understanding of what drives NPS. 
 
Previously, I was a Senior Data Analyst at Deloitte Australia. I work closely with clients to solve their complex business problems. In particular, I develop predictive models, credit risk assessment models, customer segmentation models. I also conduct geospatial analyses. Furthermore, I have extensive expertise in developing ETL solutions. 

I was a Data mining researcher at Monash University, Australia. There I developed novel machine learning algorithms to learn constraints from Airline Crew Schedules. I also developed novel machine learning algorithms, specifically extending the Naive Bayes classification algorithm. 

I was a Cognitive Scientist for Carnegie Learning Inc., USA. They develop computer software called Intelligent Tutoring Systems for teaching Mathematics. The systems have been shown to be highly effective in improving student learning. A key characteristic that set them apart from typical eLearning systems is their ability to customize the learning experience to each individual student similar to human tutors.

I completed my PhD in Computer Science from the University of Canterbury in 2006. My doctoral research was in the area of Intelligent Tutoring Systems. These systems require an encoding of the targeted area of knowledge to be taught. This encoding has to be in the form that can be interpreted by computers. My doctoral research was focused on generating the required knowledge with the assistance of an expert of the domain. As domain experts cannot be expected to learn how to encode knowledge of the domain in the formal representation, the required information was extracted from an ontology of the domain and sample problems.

Contact Information

  • Email: pramudi [at] gmail [dot] com

            Visited Countries
Quidquid latine dictum sit, altum viditur 
[Whatever is said in Latin sounds profound.]
Pramuditha Suraweera Pramuditha Suraweera 
Comments