Learn Data Skills
Beta
Vishakha Verma

Vishakha Verma

Data Scientist

ICICI | United Kingdom

Technologies

My Portfolio Highlights

My New Course

Introduction to Python

My New Course

Introduction to Python

Machine learning maverick, riding the wave of algorithms to new frontiers.

My Work

Take a look at my latest work.

course

Intermediate SQL Queries

course

Introduction to Power BI

course

Introduction to DAX in Power BI

Power BI
course

Introduction to Python

course

Intermediate SQL Queries

My Certifications

These are the industry credentials that I’ve earned.

Other Certificates

https://learn.microsoft.com/en-us/users/vishakhaverma-7499/credentials/6d59d6ceb8f1f7e9 Microsoft Azure Fundamental AZ 900

DataCamp Course Completion

Take a look at all the courses I’ve completed on DataCamp.

Courses

Introduction to Python
Introduction to Power BI
Understanding Cloud Computing
Introduction to DAX in Power BI

My Work Experience

Where I've interned and worked during my career.

ICICI Bank | Jun 2023 - Sep 2024

Junior Data Scientist

I have applied advanced machine learning (ML) techniques, statistical modeling, and optimization strategies to drive impactful business solutions. Here’s how my work has contributed: 1. Predictive Modeling for Banking - Loan Repayment Prediction: Built ML models to predict customers likely to repay their loans within 10 days, achieving an accuracy of 73%, supporting effective risk and resource management. - Business Authenticity Assessment: Designed predictive models to evaluate business legitimacy based on image data, achieving 96% accuracy, and ensuring loan approvals were granted to credible entities. - Product Purchase Prediction: Created ML models to identify customers likely to purchase bank products using call recordings, achieving an accuracy of 84%, enhancing targeted marketing strategies. 2. Voice Analytics with ML - Feature Extraction with NLP: Utilized NLP techniques using "Spacy" and "NLTK" to extract insights from call data, generating KPIs such as financial hardships, payment commitments, and call quality metrics. - Sentiment and Emotion Analysis: Conducted sentiment and emotion analysis on customer calls, enriching insights for improved service strategies. - Predictive Voice Models: Built ML models using Logistic Regression, Random Forest, and XGBoost, optimized with SMOTE, to predict repayment likelihood based on call data. Evaluated models using precision, recall, F1 score, and accuracy. 3. Image-Based Predictive Analytics - Image Data Forecasting: Developed predictive models for 100,000 manually labeled image datasets, achieving 90% accuracy with algorithms like Logistic Regression, Random Forest, and XGBoost. - Image Similarity Analysis: Implemented similarity detection techniques such as Cosine Similarity, Euclidean Distance, and Manhattan Distance to identify duplicate images, ensuring data quality. 4. Advanced Statistical Analysis - Correlation and Testing: Conducted statistical analyses to explore relationships between KPIs and target variables, such as loan repayment behavior, enabling actionable insights. - Feature Engineering: Applied advanced statistical methods to engineer features that improved the accuracy and robustness of ML models across multiple projects. Impact Summary My ML, statistical modeling, and optimization expertise have consistently empowered organizations to enhance decision-making, streamline operations, and deliver measurable results. By leveraging innovative technologies and data-driven strategies, I’ve improved accuracy, efficiency, and ROI.
Show More

MSME-TECHNOLOGY DEVELOPMENT CENTRE (PPDCAGRA) | Jul 2022 - Aug 2022

Data Science Intern

Full Stack Data Science with Python- Machine Learning and Deep Learning with real-time case study: Objective I: Classification- Student Performance Prediction • Acquired and analyzed the final grade of the students through detailed exploratory data analysis using Python libraries • Achieved an accuracy of 83% in predicting the performance after fitting the model Logistic Regression, Decision Tree Objective II: Real vs Fake Job Posting Prediction • Performed Data Pre-processing and detailed EDA to get meaningful insights from the character type dataset • Applied ML algorithms such as Random Forest and ANN to detect the difference between real and fake employment

Accenture | May 2022 - Jul 2022

Data Science Analyst

Market Mix Modeling and Optimization (RShiny Project) Objective: Developed an RShiny app to recommend better marketing strategies for the UK Life Science industry based on Market Mix Modeling and optimization techniques. 1. Optimization Code: Wrote R code to optimize budget allocation for marketing plans, providing data-driven recommendations for the most effective strategy. 2. App Deployment: Deployed the code via RShiny, creating a user-friendly interface for clients to interact with Market Mix Models and budget optimization algorithms running in the backend. 3. Research Contribution: Conducted in-depth research on RShiny, Market Mix Modeling, and optimization to create a robust and scalable solution for marketing strategy development.

My Education

Take a look at my formal education

Master's in Science in Applied Statistics and informatics in Applied Statistics and InformaticsIndian Institute of Technology, Bombay | 2023
Bachelor's in Science in Mathematical Statistics and Probability in StatisticsPatna University | 2020

About Me

Vishakha Verma

I help companies decode value from their data | Ex Data Scientist @ ICICI | Open for opportunities.

Powered by

  • Work
  • Certifications
  • Courses
  • Experience
  • Education
  • About Me
  • Create Your Data Portfolio for Free