Trishul Chowdhury

Leading NLP Research and design of data curation mechanism to convert the clinical data to machine-readable and searchable text with Optical Character Recognition (OCR) and Text summarization using NLP algorithms for Robotic Process Automation (RPA)

• Media Benchmarks and Insights: Create an actionable insights-driven summary by market/channel and media investment portfolio analysis. • Marketing mix and channel mix modeling using different open source programming language to derive optimum investment policy • Media attribution with multi-touchpoint • Regression analysis of Productivity, Effectivity, profit margin, Short-term and long-term ROI of different media channels like TV, Radio, print, Out of house marketing, Digital social, and Search.

• Marketing.Marketing mix and channel mix . • Direct email Marketing: Derived the churn rate of the customers of different clients and built models . We used a random forest algorithm and customized ANN solution to identify important KPIs and driving factors for the same which helps the business to leverage their optimum solution to retain the customer. • Social Media analysis: for different FMCG company understand the underlying consumer trends using unsupervised Natural Language processing algorithms.

• Involved in the induction of different sales data from retail team to EDW core and subsequent semantic layer for reporting purpose by mapping different order and item level metrics, incorporating complex business logic. • Apple Online Sales and Service- Enhancement and User Data Management .

Research Area – NLP, and Language model (BERT, GPT2, GPT3) Thesis Title: “Contextual Topic Modelling Using Bert As An Embedding Layer” (Liverpool John Moores University, UNITED KINGDOM)