VAISHALI Kannappan

Analytical Data Analyst with proven expertise in data visualization and statistical analysis. Skilled in using Tableau, Python, and SQL for predictive modeling and problem-solving. Adept at enhancing project outcomes through effective communication and collaboration.

• Sentiment Analysis of Customer Reviews using Yelp Review Dataset, 03/23 - 05/23, Conducted sentiment analysis on Yelp customer review dataset, applying Machine Learning (ML) techniques to classify reviews as positive, negative, or neutral., Performed Exploratory Data Analysis after pre-processing data to get valuable insights on data distribution and patterns., Built a PySpark pipeline to facilitate experimentation and deployment of Spark-based machine learning models., Performed statistical analysis and applied data visualization techniques to compare performance metrics of ML models (Naïve Bayes, Random Forest and SVM) including accuracy, precision, recall, and F1 score to choose the best performing model for classifying reviews.
• Analysis of the Effect of Domestic Coffee Consumption on Production, 03/23 - 05/23, Analyzed relationship between domestic consumption and production of coffee using inferential statistics., Analyzed impact of consumption on coffee production with statistical tools like T-Test, Chi-square test and Regression analysis., Performed production forecasting using exponential smoothing in Rapid Miner to capture future trends and seasonality.
• Design of Data Warehouse & Business Intelligence for Paul Miller Auto Group, 09/22 - 12/22, Developed a data warehouse for Paul Miller group to keep track of customer, transactional and inventory data., Designed Tableau dashboards for visualizing top selling brands, key car sales, key car trades, most preferred brands etc., Created a candidate subject area and developed detailed dimension table along with conformed dimensions for visualization., The designed data analytics solution will have the benefits of improved sales intelligence, increased competitive advantage and resale insights.
• Analysis of Airline Passenger Satisfaction using Machine Learning Techniques, 01/22 - 05/22, Built machine learning models equipped to analyze factors contributing to customer dissatisfaction with airlines., Applied CRISP-DM process on airline passenger survey dataset to gain business understanding, data preparation, evaluation, and deployment., Performed exploratory data analysis on target variables to understand factors contributing to customer satisfaction., Deployed CART (Decision Tree) and Random Forest models to classify factors contributing to customer dissatisfaction.