Data Analytics, AI, Machine Learning, BI, ETL

1. Name: Global Anti-fraud Solution (GAFS) - Fraud Analytics

Role: Manager

Description: GAFS system is a comprehensive front end portal which is used by Fraud Investigators and offers several in-built analytics,
reporting and visualization features. At the core of this system - there is a predictive model using advanced statistical and ML analytics, which
identifies fraudulent claims. It leverages historical investigation results of Claims to train the identification engine and uses 300+ claim features.
Based upon Boosted Ensemble Decision Trees, it provides feature weights and overall scoring for new or modified claims and triggers alerts
for Fraud Manager to prioritize investigations or flag suspicious claims.

Key Result Areas:
Engaged in project strategy for product documentation and diagnostic frameworks
Updated project subsystems to improve and streamline data collection
Synthesized current business intelligence data to produce reports and polished presentations highlighting findings and recommending
changes
Continuous design, improvements and deployment of ML Predictive Model based upon Boosted Ensemble Decision Trees (XGBOOST)
Managed highly imbalanced dataset, having Class undersampling, using SMOTE
Employed Machine learning algorithms for feature engineering and dimensionality reduction using Support Vector Machine, Principal
Component Analysis
Designed SmarTest which uses NLP algorithms to create an automated suite of Regression cases based on Release plan using SVM and
NLTK with inputs from JIRA for test and defects keywords and grouping
Participated in requirements meetings and data mapping sessions to understand business needs
Developed KPI metrics derived from raw company data(JIRA) to track improvements in overall project efficiency
Identified and documented detailed business rules and use cases based on requirements analysis
Coached, developed and motivated team members, providing coaching and mentoring on Python and data analysis techniques
Developed data-gathering and reporting structures from ground up and strategized methods capitalizing on system features

Key Result Areas:
Driving the analytics projects for the clients across diverse domains
Developing business solutions by using the available data and applying statistical & learning
methods
Implementing analytics project plans within preset budgets & deadlines; monitoring & reporting
on project progress
Performing analytical interpretation of technical & analytical information used to compose
detailed documentation
Facilitating end-to-end delivery of the analytics project, meeting internal and external objectives:
o Estimation, risk, scope, time, resource, quality and change management
o Value-adds, continuous process and expectation management of team
Driving the analytical scope and method for projects, including formulating and shaping the
models
Conducting computational & mathematical modelling using different packages while
coordinating with external consultants & agencies
Leading tasks which involve the use advanced modelling techniques - regression, time series,
machine learning to provide actionable results, to support new & existing products & platforms,
and to influence business decisions
Establishing scalable, efficient, automated processes for large scale data analyses, model
development, model validation and model implementation