Vishu Sablok

Data Scientist

Delhi, India7 yrs 1 mo experience

AI ML PractitionerAI Enabled

Key Highlights

Achieved 85% accuracy in predictive modeling.
Increased client revenue by $1 million.
Implemented advanced sentiment analysis using BERT.

Stackforce AI infers this person is a Data Science professional with expertise in predictive modeling and sentiment analysis in SaaS and Retail sectors.

Contact

Skills

Core Skills

Data ScienceMachine LearningNatural Language Processing

Other Skills

AWSArtificial Intelligence (AI)BERTCloud ComputingCustomer Sentiment AnalysisData AnalysisData StructuresDeep LearningDockerMLOpsNatural Language Processing (NLP)Predictive ModellingPython (Programming Language)StatisticsWeb API

About

I am a data science analyst at Accenture and have a good knowledge of the same. Looking for an opportunity to learn and gain practical exposure in the field of computer science and make a sound position in corporate world by putting all my efforts. I would like to implement my innovative ideas, skills and creativity for accomplishing the projects. As well as i'm doing some real time project on it using Python framework. Also, looking for an internship in any relevant domain

Experience

Nab

Data Scientist

Apr 2024 – Present · 1 yr 11 mos · Gurugram, Haryana, India · Hybrid

Bridgenext

Data Scientist

Aug 2022 – Mar 2024 · 1 yr 7 mos · New Delhi, Delhi, India · Remote

Accenture

Data Scientist

Dec 2018 – Jul 2022 · 3 yrs 7 mos · Delhi, India · On-site

Predictive Modelling Solution for Customer retention and acquisition in Fitness Industry
Developed a powerful predictive model for customer churn, utilizing XGBOOST and delivering an impressive 85% accuracy rate.
This model drove a remarkable 10% decrease in churn rate and a staggering $1 million increase in revenue for the client.
Deployed seamlessly on AWS ec2 instance using Docker, Web API, the model has been instrumental in boosting customer base by 20%, increasing revenue by $30,000, and significantly improving retention rates by 30%.
Customer Sentiment Analysis and Visualization for Retail Business Insights
Implemented BERT and solved the problem of identifying multiple sentiments by extracting terms and opinions from the reviews.
Analyzed the reviews to identify patterns and trends in customer satisfaction, product popularity, and key topics discussed. Tagged keywords to review to find average sentiment score for keywords

Predictive ModellingCustomer Sentiment AnalysisData AnalysisXGBOOSTBERTAWS+4