RANJITH NAGAVARAM BABU

With a Master's in Computer Science from Western University and hands-on experience as a Teaching Assistant, I have cultivated a robust blend of academic rigor and practical application. My expertise in data modeling, analytics, and Python has equipped me to transform complex data into actionable insights. At Western University, I played a pivotal role in advancing technology-driven education, empowering students to excel in Python for data analysis. My commitment to continuous learning and smart work ethic positions me to contribute effectively to dynamic organizations looking to leverage data for strategic decisions.

• Objective: Address high cancellation rates at City Hotel & Resort Hotel, optimizing revenue and room utilization.
• Tech Stack: Python, pandas, matplotlib, seaborn
• Key Achievements:Conducted comprehensive data analysis to identify key factors contributing to reduced revenue and suboptimal room utilization.
  Utilized Python libraries to load, analyze, and visualize hotel booking data, effectively communicating insights through visual representations.
  Analyzed reservation status distribution, revealing a significant proportion of canceled bookings, and visualized the data using bar charts.
  Compared reservation status across different hotel types using count plots, demonstrating varying cancellation rates between City Hotel and Resort Hotel.
  Investigated the relationship between Average Daily Rate (ADR) and cancellation rates by creating bar plots to illustrate monthly ADR for canceled reservations.
  Identified top 10 countries with the highest number of cancellations through pie chart visualization, providing actionable insights for targeted marketing strategies.
  Derived key insights such as higher cancellation rates during periods of elevated prices and longer waiting lists, as well as the prevalence of offline travel agents in reservation bookings.
  Presented data-driven recommendations to hotel management for reducing cancellation rates and optimizing room utilization, contributing to strategic decision-making.
  

• Objective: Increase the conversion rate of liability customers to personal loan customers through targeted marketing.
• Tech Stack: Python, pandas, numpy, seaborn, matplotlib
• Key Achievements:Conducted an in-depth analysis of Thera Bank's customer data to identify key factors influencing personal loan acceptance.
  Utilized Python libraries to import, clean, and analyze data from the "Bank_Personal_Loan_Modelling.xlsx" dataset.
  Visualized the distribution of customer experience using dist plots to understand its impact on personal loan acceptance.
  Created a correlation heatmap to identify significant relationships between variables, aiding in the selection of key features for targeted marketing.
  Analyzed categorical variables through count plots segmented by personal loan acceptance to identify demographic trends and preferences.
  Developed data-driven recommendations for the retail marketing department to design cost-effective campaigns with improved targeting.
  Contributed to strategic decision-making by providing actionable insights into customer behavior and loan acceptance patterns, supporting the bank's goal of enhancing customer retention and revenue growth.
  

• Objective: Explore the creation of realistic fruit images (apples) from complex genetic data using SNP data and embeddings.
• Tech Stack: Genomics, Machine Learning, Embeddings
• Key Achievements:Investigated two approaches to the creation of fruit imagery: SNP-based and embeddings-based methods.
  Used an embedding predictor to encode genetic variants linked to apple attributes into embeddings, which were then decoded into comprehensible fruit visuals.
  Compared SNP-based method with embeddings-based method, highlighting interesting differences between the two.
  Found that fruit images created with SNP data have delicate, tiny variations in shape and features, while changing the embeddings causes significant changes in the fruit's appearance.
  Advanced understanding of how various data formats affect the visual accuracy of newly generated pictures.
  Highlighted the potential and challenges of using genomic data for picture synthesis, emphasizing the importance of identifying minute modifications brought about by genetic variants.
  Expanded on existing models by utilizing experimental techniques to investigate the synthesis of pictures using genetic information.
  Contributed to the broader application of machine learning in the field of genomics by exploring different data formats within biological entity image synthesis.