This analysis was conducted using Python which applies Natural Language Processing (NLP) to analyse employee sentiment using Capgemini dataset collected from Kaggle website, which aims to uncover deeper insights beyond traditional surveys. Using techniques like VADER sentiment analysis, TF-IDF, and Word2Vec, the study trained machine learning models including Random Forest, which achieved 95% accuracy and deep learning models like CNN, which reached 86.64%. Results showed a predominance of positive feedback, highlighting themes like work-life balance and job security. The report concludes that NLP is a powerful tool for understanding employee experiences and recommends future work on real-time sentiment tracking and broader model exploration. Check out my Github for the Python Code.

This analysis presents a practical application of big data analytics using PySpark in a cloud-based environment (Google Colab). The process began with data exploration, cleaning, and merging various datasets such as customer, sales, and promotions. This prepared the foundation for comprehensive exploratory data analysis, which revealed key insights into customer demographics, sales trends by region and age group, and promotion performance. Further analysis showed strong customer retention and conversion rates, with most sales occurring in the U.S. and promotions peaking in July. The section also introduced machine learning models a Random Forest Classifier to predict promotion types with 86.7% accuracy, and a Random Forest Regressor to estimate discount percentages with high precision. Overall, this section highlights the value of PySpark for processing large datasets and extracting actionable business insights. Check out my Github for the Python Code.