Self-motivated Data-Science consultant experienced working with various cloud-based applications, data, breaking down information and applying interpretations to complex business concerns. Proficient in python, statistics, predictive and hypothetical modeling, visualization concepts and managing people & processes. Bringing almost 6 years of IT experience in strengthening company requirements
Scripting Language
ML algorithm
Statistical Methods
Working Methodology
Database Language
Data Reporting Tool
ML develop platforms
Churn Prediction
Understand the customers' past purchasing behaviors and try to predict whether the customers will come again to BigBasket platform to make any purchase? To predict it, developed an ensemble model with various models such as XGboost, Random Forest and LightGBM model with 71% of accuracy. Post the classification, the tagged churn customers are shared to marketing folks to derive specific campaigns to bring back these customers.
Orders Forecasting
To understand the daily/monthly demand at a store level, it will be easy for the operation head to fulfill the demand and supply. Developed a model using prophet by Facebook algorithm and multidimensional attributes were able to achieve 90% accuracy.
Topic Modelling
BigBasket receives customer responses across different platforms as feedback such as Facebook, In-app feedback, Playstore/Ios feedback and Instagram. To understand the customer sentiment and what's in the response? Developed a script that uses ChatGPT APIs and fine-tune the prompts accordingly to provide the issue type, nature of the issues and their sentiment. It was very useful to display the customer's opinion to the leaders.
https://www.linkedin.com/in/kumaran-sekar-847938116/